Effective dose of dental CBCT—a meta analysis of published data and additional data for nine CBCT units

Abstract

Objectives:

This article analyses dose measurement and effective dose estimation of dental CBCT examinations. Challenges to accurate calculation of dose are discussed and the use of dose–height product (DHP) as an alternative to dose–area product (DAP) is explored.

Methods:

The English literature on effective dose was reviewed. Data from these studies together with additional data for nine CBCT units were analysed. Descriptive statistics, ANOVA and paired analysis are used to characterize the data.

Results:

PubMed and EMBASE searches yielded 519 and 743 publications, respectively, which were reduced to 20 following review. Reported adult effective doses for any protocol ranged from 46 to 1073 µSv for large fields of view (FOVs), 9–560 µSv for medium FOVs and 5–652 µSv for small FOVs. Child effective doses from any protocol ranged from 13 to 769 µSv for large or medium FOVs and 7–521 µSv for small FOVs. Effective doses from standard or default exposure protocols were available for 167 adult and 52 child exposures. Mean adult effective doses grouped by FOV size were 212 µSv (large), 177 µSv (medium) and 84 µSv (small). Mean child doses were 175 µSv (combined large and medium) and 103 µSv (small). Large differences were seen between different CBCT units. Additional low-dose and high-definition protocols available for many units extend the range of doses. DHP was found to reduce average absolute error for calculation of dose by 45% in comparison with DAP.

Conclusions:

Large exposure ranges make CBCT doses difficult to generalize. Use of DHP as a metric for estimating effective dose warrants further investigation.

Introduction

The intent of this article is to review the literature on dosimetry of maxillofacial CBCT imaging in dentistry. A discussion of dose outside the context of biological harm has little relevance to patient care; therefore, this report focuses on effective dose, a quantity with direct correlations to biological risk. A number of approaches may be taken to measure dose and calculate effective dose, and some of the advantages and disadvantages of these are explored. Because children are at greater risk from exposure to ionizing radiation, child and adult doses are explored separately. Tables of published effective dose data from all protocols and equivalent dose data from standard CBCT protocols are presented together with previously unreported data for nine CBCT units. Finally, dose–area product (DAP) is contrasted with dose–height product (DHP) as a potential surrogate for estimating effective dose.

Is dose from maxillofacial radiographic imaging a relevant risk?

Dentistry had an early awareness of the dangers of exposure to ionizing radiation, as pioneers of radiographic imaging such as Edmund Kells suffered carcinoma of the hands resulting from repeated unprotected exposures during imaging of their patients.¹ We were warned about the dangers of radiography to both practitioners and patients by other pioneers such as Rollins.² But not all practitioners were convinced that there was a significant risk. In an exchange of letters published in successive February 1901 issues of the weekly predecessor to the New England Journal of Medicine, Rollins assertion that “X-light kills” was rebutted by a prominent surgeon, Ernest Codman, stating that “in careful hands, there is no danger from the use of the X-ray to the patient and very little to the operator.”³ Although the scientific debate about the level of risk associated with diagnostic imaging continues, there is substantial evidence for a cumulative dose-related response to ionizing radiation in the form of cancer developing years after initial exposure. Some of this evidence comes from the Life Span Study of atomic bomb survivors, a well-documented cohort of 105,427 people exposed to a range of doses.⁴ Analysis of these data and data from several other cohorts provides good support for an increased risk of cancer from acute exposures in a range of 10–50 mSv and chronic exposures in a range of 50–100 mSv.⁵ This has prompted support by the National Commission on Radiation Protection and Measurements for a linear extrapolation of higher dose-associated cancer risk to lower levels of exposure: “Although other dose–response relationships for the mutagenic and carcinogenic effects of low-level radiation cannot be excluded, no alternate dose–response relationship appears to be more plausible than the linear-non-threshold model on the basis of present scientific knowledge.”⁶ The Life Span Study data also indicate a significant radiation-associated increase in the risk of cancer occurring in adolescence and young adulthood.⁴ Diagnostic imaging contributes to individual and population exposures to ionizing radiation, and it has been suggested that as many as 1.5–2.0% of cancers in the USA may be related to X-ray exposure from CT imaging.⁷ Recent studies have confirmed that cancer risk extends to X-ray exposure from diagnostic imaging of the maxillofacial complex. In a Great Britain cohort of approximately 175,000 subjects who were children at the time of CT head scan exposures, cumulative doses of about 50 mGy almost tripled the risk of leukaemia and doses of about 60 mGy almost tripled the risk of brain cancer.⁸ Similar findings were seen in an Australian cohort of 10.9 million people aged 0–19 years, where a 24% increase in cancers, including brain cancers and leukaemia, were noted following CT exposure. The incidence was associated with increasing dose and young age at the time of exposure.⁹

CBCT is a form of CT that has been adapted to maxillofacial imaging and has been enthusiastically embraced by dentistry. Since its introduction in the European market in 1996 and in the US market in 2001, 15 different manufacturers have offered 24 CBCT models in the USA and many more worldwide.¹⁰ CBCT use has found its way into many aspects of general and speciality practice, including adolescent orthodontics. Public concern about this particular application has prompted questions from both patients and practitioners about safe use and best practice.¹¹ While the risk from dentomaxillofacial imaging is small for an individual, when multiplied by the large population of patients who are exposed to diagnostic imaging, radiation risk becomes a significant public health issue.

Measures of exposure to ionizing radiation

Exposure is the simplest measure of radiation dose. A variety of radiation detection devices, including ionization chambers, radiosensitive films, thermo or optical light-stimulated luminescent dosemeters, and metal oxide-semiconductor field-effect transistor devices may be used to measure ionization caused by radiation. Because calcified tissues absorb X-rays more effectively than do soft tissues and because the absorption of X-rays varies substantially with the size and shape of imaged anatomy as well as the distribution of tissues of different densities within that anatomy, exposure in air provides a limited and sometimes misleading indication of the energy imparted to different tissue types and thicknesses.

Measurement of absorbed dose in specific tissues or organs permits estimation of potential harm to a tissue of interest. Because the absorption efficiency of the radiation detection device may be different than the organ of interest, adjustment of measured values to compensate for this is necessary. For example, absorption efficiency of bone may be 2–4 times greater than that of soft tissues for average photon energies in the diagnostic spectrum. Absorbed dose is expressed in the international unit, gray and more commonly for diagnostic imaging in milligray.

Equivalent dose is absorbed dose adjusted for the attenuation characteristics of the radiation that is involved. The attenuating quality of radiation has a significant impact on biological effectiveness for cancer induction and genetic effects. Alpha particle radiation produces numerous ionization events over a short distance in comparison with the energy that is transferred by an X-ray photon over the same distance. The calculation of equivalent dose (H_T) is the product of the absorbed dose (D_T) and a radiation-weighting factor (w_R), which accounts for relative biologic effectiveness of the radiation. While w_R for α particle radiation is 20, the w_R for X-rays is 1. It is convenient, but sometimes confusing, that X-ray absorbed dose and equivalent dose are the same value. Equivalent dose is expressed in international units, sievert and again, for diagnostic imaging, more commonly in millisievert.

Although in some instances it is useful to consider the biological response to an equivalent dose for a particular tissue of interest, it is often desirable to evaluate a variety of exposures of different types and different body areas for a collective outcome such as cancer. The International Commission on Radiological Protection (ICRP) has recommended a calculation called effective dose as the preferred method for comparing risks from different exposures to ionizing radiation. Effective dose is a calculation that considers the most radiosensitive tissues and organs of the body and provides a fractional weighting reflecting the degree of sensitivity for each of those organs. Effective dose is reported in sieverts and for diagnostic imaging is more commonly expressed in millisieverts or microsieverts. Effective dose is calculated using the equation: $E={\displaystyle \sum w_{\mathrm{T}}\times H_{\mathrm{T}}}$, where E is the summation of the products of the tissue weighting factor (w_T) and the absorbed dose within that tissue H_T.¹² Because estimation of the risk of the stochastic effects of genetic mutation and cancer formation has evolved with additional data reported from observations of a variety of exposed populations, the ICRP has changed the calculation of effective dose several times. The most recent change was in 2007¹³ and is noteworthy because weights of several tissues located in and around the maxillofacial region were changed, and several other tissues within this region were added to the calculation.¹⁴ Changes in tissue weights have resulted in a 10% increase in weight of tissues located in the maxillofacial area and a 28% increase in weight after adjusting for the distribution of tissues. Newly added tissues for effective dose calculation that are entirely within the maxillofacial area include oral mucosa, salivary glands and the extrathoracic airways.

Approaches to measuring dose—do “all roads lead to Rome”?

The process of measuring equivalent dose and calculating effective dose requires a real or virtual device known as a phantom. There are numerous design variations described in the literature or commercially available that include differences in phantom size, material composition and number of dosemeter locations. While all phantoms simulate human morphology and radiation attenuation characteristics to a varying extent, the gold standard method of obtaining dosimetry for calculating effective dose utilizes an anthropomorphic phantom. Alternate techniques for calculating dose that do not use anthropomorphic phantoms include CT dose index volume (CTDIvol), dose linear product, air kerma-area product and DAP. In a previous study comparing an anthropomorphic phantom and a standard acrylic cylinder with a single ion chamber used to calculate (CTDIvol), we demonstrated that the standard acrylic cylinder underestimates effective dose by 38–62%.¹⁵ This underestimation is in part owing to the failure to account for scatter dose to tissues outside of the scan region. Kerma-area product is another method that has recently been used to calculate dose.¹⁶ Values reported in the referenced study underestimate effective dose by 90–300% when compared with effective dose calculated from anthropomorphic phantom data.¹⁷ DAP has also been suggested as a simple approach for calculating dose. However, our experiments with the SCANORA^® 3D (Soredex, Helsinki, Finland) unit revealed an approximately three-fold change in effective dose between various locations of the small field of view (FOV) with no change in DAP.¹⁸ By contrast, anthropomorphic phantoms made from materials that have similar X-ray attenuation characteristics as human tissue and have multiple dosemeters allow for accurate measurement of absorbed dose. In a recent study, we confirmed that an anthropomorphic phantom using bone equivalent material in place of a human skeleton could provide reliable measures of effective dose.¹⁹ Virtual phantoms and Monte Carlo simulation of exposure have been used in assessment of organ dose and effective dose for a number of studies. Dose correspondence with anthropomorphic phantom studies is dependent on the virtual phantom that is used as well as imaging geometry and technical parameters. In one study comparing four different phantoms developed from CT data, a 70% difference in effective dose was noted for a large FOV scan depending phantom choice.²⁰ When a cephalometric analysis was performed on ICRP adult male and female phantoms, a 17° downward rotation difference of the Frankfort plane was seen in the female. This downward rotation of the chin effectively moves the thyroid closer to the radiation field and results in an average of a 3.7-fold increase in female thyroid dose for four large FOVs that were evaluated in one study.²¹ Similar increases are seen in oesophageal dose in this phantom. Although the ICRP phantoms were developed for general dosimetric applications, the inability to adjust phantom posture to establish a Frankfort horizontal plane or any orientation other than the original orientation used to acquire axial CT slices limits the accuracy of these phantoms in simulations of dental CBCT diagnostic protocols. Although virtual phantoms with Monte Carlo simulation hold a great deal of promise, further attention must be given to virtual phantom development consistent with positioning for standard dental diagnostic protocols before this becomes a reliable replacement for anthropomorphic phantoms.

Even anthropomorphic phantom systems are subject to variations in dosimetry from a variety of sources. Intended differences in subject size and organ location are seen when comparing child and adult phantoms. Child phantom effective doses are approximately 36% greater than adult phantom doses for the same imaging protocol.²² This difference is largely related to the proximity of the thyroid gland to the lower border of the mandible. Because of reduced distance between the thyroid and the mandible in a child, direct exposure of the thyroid is more likely and the intensity of scatter radiation from jaw structures to the thyroid is greater.²²

Use and abuse of effective dose

Effective dose was developed to provide a measure of stochastic risks from exposures to low doses of ionizing radiation. While developed for use in radiation protection, it should not be applied to estimations of individual patient risks. There are several reasons for this. Foremost, effective dose represents risk to a reference subject who is an average of characteristics, including age, gender and genetic radiation sensitivity. Another reason is that effective dose estimation is subject to numerous sources of uncertainty.²³ Among these is the need to extrapolate stochastic outcomes associated with higher doses to the low doses associated with dental diagnostic imaging. X-ray beam shape and imaging geometry for organs partially in or just outside the beam is also estimated to account for as much as ±40% variation. Despite these limitations, effective dose is a useful metric for comparing alternative imaging modalities or examination protocols in terms of relative risk. Employing the same methods of dose measurement for different examinations, units or protocols, we can evaluate which produces a greater or lower risk. The caveat that this risk may be greater or less depending on individual patient characteristics makes the comparison no less valid.

Biological parameters that influence dose

A number of physical and biological parameters influence individual dose and risk. Age has a significant impact on both. Children are physically smaller, which places peripherally located brain and thyroid tissues closer to the dental area that is being imaged. Even if not directly exposed, these organs will receive increased scatter radiation with increased proximity to the location of the scanned volume. But children are not simply small adults. They are also at increased risk from any exposure to ionizing radiation owing to cellular growth and organ development, which increases radiosensitivity of tissues. In conjunction with a longer life expectancy in which cancer can develop, children may be two times or more sensitive to radiation carcinogenesis than are mature adults.^24,25 Physical differences associated with gender are also associated with differences in risk. Females are at significant risk for breast cancer, while males are not. Females are at risk for ovarian cancer, while males are at risk for prostate cancer. Because these organs are distant from the maxillofacial area, gender differences do not impact dose and risk estimation for maxillofacial imaging.

Technical parameters that influence dose

Receptor technology and field of view

Multiple technical parameters influence patient dose. Two types of receptor technologies are used to acquire image data. Image intensifiers utilize a round receptor and produce a spherical FOV. Square or rectangular flat panel detectors are incorporated in many CBCT units, and these produce a cylindrical FOV. In general, the cylindrical field is more efficient at capturing the anatomy of the maxillofacial complex when the top of the field includes the temporomandibular joint areas. A cylindrical volume diameter, which captures both temporomandibular joint and chin anatomy will require a spherical volume diameter that is approximately 25% larger to cover the same anatomy.

mAs

X-ray tube current (mA) and exposure time (s) are directly proportional to dose when other factors remain constant. The product of mA and s (mAs) is also directly proportional. For instance, doubling mAs doubles dose. It should be mentioned that exposure time may not be the same value as scanning time. Some CBCT units produce continuous output of radiation during scanning. For these units, scan time is equal to exposure time. However, most detectors are unable to record X-ray exposure during the period when the image detector integrates the X-ray energy absorbed in individual receptor pixels and transfers this signal to the computer. Continued X-ray exposure during signal integration contributes to patient dose but adds nothing to image formation. To eliminate this unnecessary patient exposure, many CBCT units utilize a pulsed X-ray source, where X-ray emission is intermittently turned off during the image acquisition process.

kVp and beam filtration

X-ray beam quality has long been noted as a factor associated with patient dose from diagnostic imaging. Increasing filtration of the X-ray beam reduces patient exposure to lower energy X-ray photons that are more likely to contribute to patient dose without contributing to image formation.²⁶ While higher beam energies (kV) are associated with loss of contrast in film-based imaging, digital imaging affords the possibility of post-acquisition contrast enhancement. Use of 0.4 mm of additional copper filtration in conjunction with increased kVp was demonstrated to reduce patient dose by an average of 43% with one unit.¹⁰ In a study of a different manufacturer's unit, the effective dose for a standard exposure with an 8 × 8-cm FOV was reduced 57% using 0.5 mm of additional copper filtration when compared with the dose produced by an earlier version of the unit.^15,19

Resolution

In order to maintain adequate signal-to-noise level, exposure must be increased as the voxel size is reduced to create higher resolution images. This can take place as an increase in mA or an increase in the number of basis images that are acquired. With some CBCT units, this choice is under operator control, but, at other times, the unit dictates which exposure factors may be used with different resolutions. Automatic doubling of dose when switching from standard to high resolution has been reported for one unit.²²

Methods and materials

Systematic review of literature on CBCT and effective dose

A systematic review of the literature concerning CBCT dosimetry in the maxillofacial region was performed. A PubMed (MEDLINE) database (National Library of Medicine, NCBI) search was performed on 11 November 2013 and updated on 26 May 2014. An EMBASE search was also performed on 26 May 2014 by a senior librarian at the University of North Carolina Health Science Library, Chapel Hill, NC. The strategy to search for publications in English language indexed in the MEDLINE database was as follows: {“Cone-Beam Computed Tomography” [(Mesh)] OR CBCT (tw) OR CBVT (tw) OR Cone beam computed tomography (tw) OR Cone beam volumetric tomography (tw)} AND [“Radiation Monitoring” (Mesh) OR “Radiation dosage” (Mesh) OR absorbed dos* (tw) OR equivalent dos* (tw) OR effective dos* (tw) OR dosimetry (tw)]. The strategy to search for publications in English language indexed in the EMBASE database was as follows: (“cone beam computed tomography”/exp OR “cone beam computed tomography scanner”/exp OR CBCT:ti,ab OR CBVT:ti,ab OR “Cone beam computed tomography”:ti,ab OR “Cone beam volumetric tomography”:ti,ab) AND (“radiation monitoring”/exp OR “radiation dose”/exp OR “dosimetry”/exp OR “absorbed dose”:ti,ab OR “absorbed doses”:ti,ab OR “absorbed dosage”:ti,ab OR “absorbed dosages”:ti,ab OR “equivalent dose”:ti,ab OR “equivalent doses”:ti,ab OR “equivalent dosage”:ti,ab OR “equivalent dosages”:ti,ab OR “effective dose”:ti,ab OR “effective doses”:ti,ab OR “effective dosage”:ti,ab OR “effective dosages”:ti,ab). The PubMed search yielded 519 articles and the EMBASE search yielded 743 articles. Inclusion criteria for this review were all articles published in English in the scientific literature related to CBCT dosimetry in the maxillofacial region. Only articles utilizing tissue weights from the 2007 ICRP recommendations for calculating effective dose were included. The articles had to include pertinent information regarding the scanner used, FOV size and location, exposure technique, phantom type and dosemeter used. Articles not meeting the inclusion criteria were excluded after downloading the references in EndNote^® (Thompson Reuters, Rochester, NY) and reviewing the abstracts. 65 articles were initially included and the PDF of the articles were downloaded in EndNote and reviewed in more detail. 43 more articles were excluded leaving 22 articles. Data from these studies were placed in a spreadsheet for analysis. Data identified as outliers led to the identification of methodological errors resulting in the removal of two studies and reassessment of the data. Data from 20 studies and additional unpublished data otherwise meeting the inclusion criteria were ultimately tabulated and presented in this article. Reasons for exclusion of the 43 manuscripts are catalogued in Table 1.

Table 1.

Reasons for exclusion of 45 of 65 citations resulting from literature search

Reason for exclusion	Number
Dosimetry based on 1990 International Commission on Radiological Protection calculation of effective dose rather than 2007 recommendations	6
Image-guided radiation therapy device or other non-dental unit	10
Incomplete dosimetry—multiple organs in the head and neck area have been omitted including remainder tissues	6
Not a dosimetry article, a review article, letter or other secondary source	7
Technique article that does not provide information about specific or identifiable units	3
Effective dose not reported	9
Animal study or anatomy other than maxillofacial area	2
Foreign language with English abstract	1
Other—reported results violate dose/mAs linearity by ×3 indicating an unrecognized calculation error	1

Additional dosimetry

Previously unreported data are included for nine additional CBCT units: 3D Accuitomo (J Morita, Osaka, Japan), CS 9000 (Carestream Dental, Atlanta, GA), CS 9300 (Carestream Dental), Orthophos XG 3D (Sirona Dental Systems, Bensheim, Germany), Galileos Comfort Plus (Sirona Dental Systems), ProMax Mid (Planmeca Oy, Helsinki, Finland), NewTom VGi (Cefla Dental Group, Imola, Italy) and OP 300 Maxio (Instrumentarium, Helsinki, Finland). Scanning protocols for these units are found in Table 2.

Table 2.

Units and protocols used to produce unreported dosimetry included in this manuscript

CBCT unit	Manufacturer	Phantom type	Protocol	Field of view
3D Accuitomo	J Morita (Osaka, Japan)	Atom child, adult	Standard, high fidelity, high resolution, high speed/360°, 180° scans/child, average adult, large adult	Large, medium, small
CS 9000	Carestream Dental (Atlanta, GA)	Rando adult	Standard	Small
CS 9300	Carestream Dental	Atom child, adult	Standard	Medium, small
Orthophos XG 3D	Sirona Dental Systems (Bensheim, Germany)	Atom child, adult	Standard, high definition, endo/child, teen, small adult, average adult, large adult	Small
Galileos® Comfort Plus	Sirona Dental Systems	Atom child, adult	Standard, high definition/child, teen, small adult, average adult, large adult	Medium, small
ProMax® Mid	Planmeca Oy (Helsinki, Finland)	Atom child, adult	Normal, low dose, high definition/child, adolescent, small adult, average adult, large adult	Medium
NewTom VGi	Cefla Dental Group (Imola, Italy)	Rando adult	Standard, high resolution	Medium, small
NewTom 3G	Cefla Dental Group	Atom child	Standard	Large, medium
OP 300 Maxio	Instrumentarium (Helsinki, Finland)	Atom child, adult	Standard, low dose, high definition, endo/child, small adult, regular adult, large adult	Medium, small

Following previously published protocols that utilize 24 dosemeters placed in and on anthropomorphic phantoms, dosimetry was acquired for standard imaging protocols and additional protocols when available. Head and neck phantoms replicating the radiation attenuation characteristics of human tissues and anatomy were used.^15,22 The child phantom (Atom Model 706 HN; CIRS Inc., Norfolk, VA) simulated characteristics of a 10-year-old child. An adult phantom simulated an average adult male (Atom Max Model 711 HN; CIRS Inc.).

Optically stimulated luminescent dosemeters (nanoDot™, Landauer, Inc., Glenwood, IL) were cleared of ambient charge prior to use, using a minimum of 12 h of exposure to light from a florescent tube, dental radiographic film view box. Dosemeters were read to record residual baseline energy level using a portable reader (MicroStar; Landauer, Inc.). The reader was calibrated before use using a set of 80-kVp reference dosemeters supplied by the manufacturer. After adjusting for individual dosemeter energy sensitivity, photon counts were converted to dose and automatically recorded in a database by the reader. After placement in a phantom and exposure to the CBCT scan, dosemeters were read three times with the reader. The modal value of the three readings was selected as the dose of the dosemeter. Doses were exported from the database as an Excel^® (Microsoft, Redmond, WA) spreadsheet and adjusted for response to the estimated mean energy of the X-ray beam using a third-order polynomial calibration curve derived from side-by-side comparison of recorded doses from an ion chamber and optical stimulated luminescent dosemeters over a range of 80–120 kVp using an adjustable kVp source. Beam energy adjustments ranged from 0.97 for an 84-kVp source (mean kV = 56) to 0.78 for a 120-kVp source (mean kV = 80). 2–20 exposures were utilized for each dosemeter run to provide a more reliable measure of radiation in the dosemeters. Smaller FOVs require more exposure repetition because more dosemeters are outside the field of direct exposure and absorb only small quantities of scatter radiation. For every scan, a scout view was also acquired. Dosemeter values were divided by the number of scans to determine the “exposure per examination” for each dosemeter.

Absorbed dose for a tissue or organ used in the estimation of effective dose was calculated by averaging doses for dosemeters located within that tissue and are reported in micrograys (µGy).²² In instances where tissues were not fully contained within the head and neck area, an estimation of the proportion of this tissue within this area was used to calculate organ absorbed dose. For skin surface, lymph nodes and muscle, an estimate of 5% was used. For the oesophageal tract, an estimate of 10% was used. Calculations for bone surface and bone marrow were adjusted for calvarial, jaw or spine location as well as phantom type (child, adult) using estimations of Underhill et al²⁷ for bone distribution and Christy²⁸ for marrow distribution. For bone, a correction factor based on experimentally determined mass energy attenuation coefficients for bone and muscle irradiated with monoenergetic photons was applied. Effective beam energy estimated to be two-thirds of the peak beam energy of the CBCT unit was used to determine bone/muscle attenuation ratios. A linear fit (R² = 0.996) of ratios from 40 to 80 kV from published data²⁹ was used to calculate bone/muscle ratio for the CBCT unit kVp setting. Calculated values provided bone/muscle attenuation ratios from 3.46 at 54.0 kV (84 kV peak) to 1.97 at 80 kV (120 kV peak) for the units and protocols investigated in this study. The products of absorbed dose and the percentage of a tissue or organ irradiated in the CBCT examination were used to calculate equivalent dose in microsieverts (µSv). Effective dose (E), expressed in µSv, was calculated using ICRP 2007 tissue weighting factors.¹³

Data analysis

Effective doses for various exposure parameters and protocols are reported in a tabular format together with equivalent doses from standard or default CBCT imaging if these were included in the publication. Means and variance for data grouped by child or adult phantom and small, medium and large FOVs are reported in summary tables. For this manuscript, small FOVs are defined as any field with a height ≤10 cm. Medium FOVs include a range of volume heights from 10 to 15 cm. Large FOVs have volume heights >15 cm. ANOVA is used to distinguish differences in equivalent doses or effective doses owing to the variables of phantom and FOV. An additional ANOVA examines the effect of maxillary or mandibular position on effective dose for small FOVs. An α level of 0.05 was selected for statistical significance.

The product of salivary gland dose and the dimensions of FOV (H × W) are used as a surrogate for DAP to calculate conversion coefficients for effective dose estimation. Similarly, the product of salivary gland dose and volume height alone (DHP) is calculated to investigate the possible use of this metric for calculating effective dose conversion coefficients. This is analogous to the product of CTDI and scan length (dose–length product) as a dose metric in CT imaging. The absolute error between the estimated effective dose derived from DAP and the phantom dose measurement was compared with absolute error between DHP-derived dose and phantom measurement in a matched pairs analysis.

Results

Table 3 displays exposure parameters and doses for an adult phantom and large FOVs. Reported effective doses from standard protocols ranged from 46 to 916 µSv. Table 4 lists doses for medium FOVs. Reported effective doses from standard protocols ranged from 47 to 560 µSv. Tables 5 and 6 provide doses for maxillary and mandibular small FOVs, respectively. Standard protocol doses ranged from 5 to 140 µSv for maxillary views and from 18 to 488 µSv for views including the mandible. Table 7 catalogues temporomandibular joint FOVs. Table 8 combines large and medium FOVs for child phantoms. Doses from standard protocols ranged from 39 to 430 µSv. Tables 9 and 10 list child doses for maxillary and mandibular small FOVs, respectively. Maxillary effective doses from standard protocols ranged from 16 to 177 µSv, while FOVs including the mandible ranged from 24 to 331 µSv. Reported adult effective doses for any protocol in Tables 3–6 ranged from 46 to 1073 µSv for large FOVs, 9–560 µSv for medium FOVs and 5–652 µSv for small FOVs. Child effective doses from any protocol in Tables 7–10 ranged from 13 to 769 µSv for large or medium FOVs and 5–582 µSv for small FOVs. Although standard protocols were the focus of this study, the included reports and additional data provided a total of 41 large FOV protocols, 81 medium FOV protocols and 249 small FOV protocols for adult phantom imaging. For child imaging protocols, the totals were 8, 35 and 103 for large, medium and small FOVs, respectively.

Table 3.

Adult phantom equivalent and effective doses for standard or default exposures for large field of view (FOV) CBCT units (>15 cm height)

Unit name	Manufacturer	FOV size H × W (cm)	kVp	mAs	Effective dose (µSv)	Bone marrow (µSv)	Bone surface (µSv)	Skin (µSv)	Oesophagus (µSv)	Brain (µSv)	Thyroid (µSv)	Salivary glands (µSv)	Remainder (µSv)	Study
ProMax Mid-stiched	Planmeca Oy (Helsinki, Finland)	16 × 16	90	(108, 127, 145) 271, 325, 380 “361, 433, 506”	(95, 112, 128) 223, 283, 339 “304, 365, 426”	345	1428	235	145	2436	345	5049	189	Current study
SkyView®	Cefla Dental Group (Imola, Italy)	17 × 17	90	51.5	87	134	125	58	–	719	474	1582	224	Pauwels et al30
iCAT NG	Imaging Sciences (Hatfield, PA)	17 × 23	120	18.7	46	40	160	60	20	660	50	850	86	Morant et al21
3D eXam®	Imaging Sciences	17 × 23	120	37	156	–	–	–	–	–	–	–	–	Rottke et al31
3D eXam	Imaging Sciences	17 × 23	120	18.5	72	–	–	–	–	–	–	–	–	Schilling and Geibel32
iCAT® FLX	Imaging Sciences	17 × 23	120	18.5, 37	69, 136	84	202	50	39	668	301	1293	195	Ludlow and Walker22
iCAT NG	Imaging Sciences	17 × 23	120	18.5, 37	67a, 129	139	277	39	15	888	159	1158	165	Grunheid et al33
iCAT NG	Imaging Sciences	17 × 23	120	18.5	74	147	294	52	33	950	183	1250	186	Ludlow and Ivanovic15
iCAT NG	Imaging Sciences	17 × 23	120	18.5	78	–	–	–	–	–	–	–	–	Davies et al34
iCAT NG	Imaging Sciences	17 × 23	120	37	182	–	–	–	–	–	–	–	–	Roberts et al 200935
Alphard VEGA	Asahi Roentgen (Kyoto, Japan)	18 × 20	80	68, 102	123, 183	427	1982	292	28	1530	533	4090	532	Kim et al36
CS 9500	Carestream Dental (Atlanta, GA)	18 × 20	90	108	136	206	215	92	–	1205	585	2676	380	Pauwels et al30
CS 9500	Carestream Dental	18 × 20	80, 85, 90	86.4, 108, 108	93, 163, 260	218	747	123	131	1640	835	2645	389	Ludlow10
CS 9500	Carestream Dental	18 × 20	90	108	151	–	–	–	–	–	–	–	–	Rottke et al31
CB Mercuray	Hitachi (Tokyo, Japan)	19 × 19	100, 120	100, 150	569, 1073	692	3211	389	393	3967	6333	5467	828	Ludlow and Ivanovic15
CB Mercuray	Hitachi	19 × 19	120b	150b	916	1726	8019	846	365	10,100	3200	13,900	1976	Librizzi et al37
CB Mercuray	Hitachi	19 × 19	80, 100, 100, 120	100, 100, 150, 150	256, 466, 683, 932	–	–	–	–	–	–	–	–	Jadu et al38
Iluma	Imtec (Ardmore, OK)	19 × 19	120b	76, 152	94, 157	–	–	–	–	–	–	–	–	Vassileva and Stoyanov16
Ilumina	Imtec	19 × 19	120	20	98	161	745	82	50	1267	350	1661	248	Ludlow and Ivanovic15
Ilumina	Imtec	19 × 19	120b	152b	498	834	3869	421	233	6267	1733	8400	1265	Ludlow and Ivanovic15
NewTom 3G	Cefla Dental Group	19 × 19	110	8.1	68	125	581	62	57	700	333	956	140	Ludlow and Ivanovic15
NewTom 9000	Cefla Dental Group	19 × 19	110	Auto	95	78	190	60	60	180	775	1550	273	Qu et al39
DCT PRO	VATECH (Seoul, Korea)	19 × 20	90	105	254	391	1260	160	143	1740	1895	3210	518	Qu et al40

H, height; kVp, kilovolt peak; mAs, milliampere per second; W, width.

Bold values represent parameters used to produce standard or default scans.

^aAverage of 0.3 and 0.4 voxel scan data.

^bInitial manufacturer recommended exposure—subsequently reduced.

Table 4.

Adult phantom equivalent and effective doses for standard or default exposures for medium field of view (FOV) CBCT units (10–15 cm height)

Unit name	Manufacturer	FOV size H × W(cm)	kVp	mAs	Effective dose (µSv)	Bone marrow (µSv)	Bone surface (µSv)	Skin (µSv)	Oesophagus (µSv)	Brain (µSv)	Thyroid (µSv)	Salivary glands (µSv)	Remainder (µSv)	Study
3D Accuitomo 170	J Morita (Osaka, Japan)	10 × 10	90	45, 79, 87.5, 154	132, 232, 257, 453	267	831	152	140	176	1498	5487	776	Current study
Alphard VEGA	Asahi Roentgen (Kyoto, Japan)	10 × 10	80	68, 136	85, 184	217	1009	147	43	496	872	6233	788	Kim et al36
CB Mercuray	Hitachi (Tokyo, Japan)	10 × 10	120	150	407	466	2161	344	110	2950	1300	9006	1355	Ludlow and Ivanovic15
CB Mercuray	Hitachi	10 × 10	80, 100, 120	100, 100, 150	148, 261, 421	–	–	–	–	–	–	–	–	Jadu et al38
Alphard VEGA	Asahi Roentgen	10 × 10	80	68, 136	69, 146	196	907	308	17	1019	403	4544	659	Kim et al36
CB Mercuray	Hitachi	10 × 10	80, 100, 120	100, 100, 150	60, 97, 145	–	–	–	–	–	–	–	–	Jadu et al38
CS 9300	Carestream Dental (Atlanta, GA)	10 × 10	90	25	76	57	208	44	37	123	359	1855	257	Current study
3D Accuitomo 170	J Morita	10 × 14	90	45, 79, 87.5, 154	138, 242, 269, 473	286	897	171	150	292	1372	5861	843	Current study
3D Accuitomo 170	J Morita	10 × 14	90	87.5	188	–	–	–	–	–	–	–	–	Theodorakou et al41
NewTom VG	QR (Verona, Italy)	10 × 15	110	10.4	83	115	163	50	–	251	354	1690	281	Pauwels et al30
iCAT NG	Imaging Sciences (Hatfield, PA)	10 × 16	120	10, 18.5	32, 53	30	120	50	40	190	70	1150	98	Morant et al21
DCT PRO	VATECH (Seoul, Korea)	10 × 16	90	105	249	170	550	110	165	290	2700	3780	556	Qu et al40
NewTom VG	Cefla Dental Group (Imola, Italy)	11 × 15	110	Auto	81	–	–	–	–	–	–	–	–	Theodorakou et al41
iCAT NG	Imaging Sciences	11 × 16	120	10, 18.5	36, 58	40	150	60	40	310	80	1230	110	Morant et al21
iCAT® FLX	Imaging Sciences	11 × 16	90/120	6/10, 18.5, 37	9/43, 79, 159	79	176	41	43	238	353	1859	256	Ludlow and Walker22
CS 9300	Carestream Dental	11 × 17	90	25.6, 51.5	101, 204	210	735	158	116	820	930	4445	626	Current study
NewTom VGi	QR	12 × 15	110	6.2	103	100	269	113	54	878	477	2076	301	Current study
3D Accuitomo 170	J Morita	12 × 17	90	45, 79, 87.5, 154	154, 260, 325, 532	330	1052	295	200	446	1951	5843	894	Current study
3D Accuitomo 170	J Morita	12 × 17	90	87.5	216	–	–	–	–	–	–	–	–	Theodorakou et al41
OP300 Maxio	Instrumentarium (Helsinki, Finland)	13 × 15	90	29, 36, 45, 72	66, 82, 102, 164	122	414	72	52	180	427	2096	342	Current study
i-CAT Classic	Imaging Sciences	13 × 16	120	18.5	69	95	1450	149	30	567	267	1450	53	Ludlow and Ivanovic15
iCAT NG	Imaging Sciences	13 × 16	120	18.5	87	105	211	82	45	808	283	1836	67	Ludlow and Ivanovic15
iCAT NG	Imaging Sciences	13 × 16	120	10, 18.5	40, 66	50	200	80	40	590	80	1270	123	Morant et al21
iCAT NG	Imaging Sciences	13 × 16	120	18.5	83	116	124	54	–	375	355	1830	260	Pauwels et al30
3D eXam®	Imaging Sciences	13 × 16	120	18.5	107	–	–	–	–	–	–	–	–	Schilling and Geibel32
iCAT FLX	Imaging Sciences	13 × 16	90/120	6/10, 18.5, 37	11/54, 85, 171	85	185	74	48	380	405	1898	265	Ludlow and Walker22
iCAT NG	Imaging Sciences	13 × 16	120	18.5	77	–	–	–	–	–	–	–	–	Davies et al34
iCAT NG	Imaging Sciences	13 × 16	120	18.5	111	–	–	–	–	–	–	–	–	Roberts et al35
iCAT NG	Imaging Sciences	13 × 16	120	18.5	82	–	–	–	–	–	–	–	–	Theodorakou et al41
SCANORA® 3D	Soredex (Helsinki, Finland)	13.5 × 14.5	85	48	68	86	94	55	–	255	296	1568	221	Pauwels et al30
CS 9300	Carestream Dental	13.5 × 17	90	45.2	184	159	572	150	103	789	907	4006	575	Current study
Iluma Elite	Imtec (Ardmore, OK)	14 × 21	120	76	368	660	667	277	–	3415	1230	7225	1034	Pauwels et al30
Alphard VEGA	Asahi Roentgen	15 × 15	80	85, 153	158, 288	609	2827	421	36	2834	821	6931	892	Kim et al36
CB Mercuray	Hitachi	15 × 15a	120	150	548	874	5110	569	100	8950	1075	8760	337	Librizzi et al37
CB Mercuray	Hitachi	15 × 15a	120	150	560	940	4360	641	177	5933	1700	10,561	379	Ludlow and Ivanovic15
CB Mercuray	Hitachi	15 × 15a	100	96	227	485	487	225	50	3032	473	3673	586	Lukat et al42
CB Mercuray	Hitachi	15 × 15a	80, 100, 120	100, 100, 150	153, 275, 435	–	–	–	–	–	–	–	–	Jadu et al38
Galileos Comfort	Sirona Dental Systems (Bensheim, Germany)	15 × 15a	85	21, 42	70, 128	82	382	40	37	267	233	1606	57	Ludlow and Ivanovic15
Galileos Comfort	Sirona Dental Systems	15 × 15a	85	28	84	82	83	55	–	124	380	2104	292	Pauwels et al30
Galileos Comfort	Sirona Dental Systems	15 × 15a	85	21, 42	51, 95	–	–	–	–	–	–	–	–	Rottke et al31
Galileos Comfort Plus	Sirona Dental Systems	15 × 15a	98	8, 10, 12 (20, 25, 30)	38, 47, 56 (106, 130, 154)	43	135	35	29	165	245	1011	144	Current study
NewTom VGi	QR	15 × 15	110	7.8	97	107	299	101	51	1002	440	1852	269	Current study
NewTom VGi	QR	15 × 15	110	8.8	194	186	184	98	–	605	2045	2855	436	Pauwels et al30

H, height; kVp, kilovoltage peak; mAs, milliampere per second; W, width.

Bold values represent parameters used to produce standard or default scans.

^aSpherical FOV.

Table 5.

Adult phantom equivalent and effective doses for standard or default exposures for small field of view (FOV) CBCT units (<10 cm height) maxillary views

Unit name	Manufacturer	FOV size H × W(cm)	kVp	mAs	Effective dose (µSv)	Bone marrow (µSv)	Bone surface (µSv)	Skin (µSv)	Oesophagus (µSv)	Brain (µSv)	Thyroid (µSv)	Salivary glands (µSv)	Remainder (µSv)	Study
Veraviewepocs 3D	J Morita (Osaka, Japan)	4 × 4a	80	47.5	21	–	65	17	–	1	1	1890	–	Al-Okshi et al43
3D Accuitomo 170	J Morita	4 × 4a	90	87.5	32	–	–	–	–	–	–	–	–	Theodorakou et al41
CS 9000	Carestream Dental (Atlanta, GA)	4 × 5a	70	107	5	5	22	47	–	30	–	130	20	Current study
CS 9000	Carestream Dental	4 × 5b	70	107	10	12	49	41	2	60	–	220	38	Current study
CS 9000	Carestream Dental	4 × 5a	70	107	19	21	27	25	–	18	30	523	74	Pauwels et al30
CS 9000	Carestream Dental	4 × 5a	70	85.6	24	–	–	–	–	–	–	–	–	Theodorakou et al41
ProMax 3Dc	Planmeca Oy (Helsinki, Finland)	4 × 5a	84	120	10	–	40	12	–	1	13	650	–	Al-Okshi et al43
3D eXam®	Kavo (Bieberach, Germany)	4 × 16	120	18.5, 37	33, 68	–	–	–	–	–	–	–	–	Schilling and Geibel32
Alphard VEGA	Asahi Roentgen (Kyoto, Japan)	5 × 5a	80	102, 153	20, 22	43	199	73	3	115	99	364	98	Kim et al36
Alphard VEGA	Asahi Roentgen	5 × 5b	80	102, 153	20, 25	41	192	133	6	169	104	494	105	Kim et al36
CS 9300	Carestream Dental	5 × 5a	84	60, 100	35, 59	40	170	13	15	56	146	1884	219	Current study
CS 9300	Carestream Dental	5 × 5b	84	60, 100	48, 80	65	290	14	21	80	191	2399	303	Current study
OP300 Maxio	Instrumentarium (Helsinki, Finland)	5 × 5a	90	11.7, 14.7, 18.7, 23.4	12, 16, 20, 25	40	128	17	5	20	42	318	68	Current study
PaX-Uni3D	VATECH (Seoul, Korea)	5 × 5a	85	120	44	47	49	55	–	28	209	1073	146	Pauwels et al30
Orthophos XG	Sirona Dental Systems (Bensheim, Germany)	5 × 5.5a	85	36, 51, 66, (72, 86, 101)	21, 30, 39, (45, 53, 60)	26	120	16	8	28	76	896	108	Current study
Orthophos XG	Sirona Dental Systems	5 × 5.5b	85	36, 51, 66, (72, 86, 101)	25, 36, 47, (58, 70, 81)	33	156	18	10	31	90	1036	131	Current study
ProMax 3Dc	Planmeca Oy	5 × 8	84	192	131	98	341	70	19	86	333	3865	514	Qu et al19
3D Accuitomo 170	J Morita	5 × 10	90	87.5	54	112	112	62	–	189	148	2138	85	Pauwels et al30
3D Accuitomo 170	J Morita	5 × 10	90	45, 79, 87.5, 154	58, 102, 113, 198	224	668	102	27	147	249	1951	382	Current study
CS 9300	Carestream Dental	5 × 10	90	25	56	36	141	8	13	52	116	1264	171	Current study
3D Accuitomo 170	J Morita	5 × 14	90	45, 79, 87.5, 154	70, 123, 136, 240	256	767	106	34	201	299	2477	474	Current study
3D Accuitomo 170	J Morita	5 × 14	90	87.5	70	–	–	–	–	–	–	–	–	Theodorakou et al41
3D Accuitomo 170	J Morita	5 × 17	90	45, 79, 87.5, 154	68, 119, 132, 232	240	721	85	36	205	298	2452	460	Current study
Pan eXam Plus 3D	Kavo	6 × 4	90	23, 49	40, 79	–	–	–	–	–	–	–	–	Schilling and Geibel32
3D Accuitomo 170	J Morita	6 × 6a	90	27, 45, 52.5, 79, 87.5, 154	19, 32, 37, 56, 62, 109	107	325	107	18	104	163	1126	210	Current study
3D Accuitomo 170	J Morita	6 × 6b	90	27, 45, 52.5, 79, 87.5, 154	20, 33, 39, 58, 65, 114	104	326	78	15	123	140	1068	250	Current study
NewTom VGi	QR (Verona, Italy)	6 × 6b	110	70.1	140	191	486	101	47	1028	355	2763	478	Current study
NewTom VGi	QR	6 × 6a	110	65	131	173	442	132	42	982	332	2612	443	Current study
OP300 Maxio	Instrumentarium	6 × 8	90	11.7, 14.7, 18.7, 23.4	25, 31, 40, 50	87	275	27	9	46	79	621	134	Current study
Pan eXam Plus 3D	Kavo	6 × 8	90	47, 79	79, 125	–	–	–	–	–	–	–	–	Schilling and Geibel32
iCAT® FLX	Imaging Sciences (Hatfield, PA)	6 × 16	90/120	6/10, 18.5, 37	4/20, 32, 65	32	73	11	12	131	101	719	119	Ludlow and Walker22
iCAT NG	Imaging Sciences	6 × 16	120	18.5, 37	32, 60	–	–	–	–	–	–	–	–	Davies et al34
iCAT NG	Imaging Sciences	6 × 16	120	10, 18.5	22, 35	–	–	–	–	–	–	–	–	Morant et al21
iCAT NG	Imaging Sciences	6 × 16	120	18.5, 37	37, 68	–	–	–	–	–	–	–	–	Roberts et al35
iCAT NG	Imaging Sciences	6 × 16	120	18.5	33	–	–	–	–	–	–	–	–	Theodorakou et al41
iCAT NG	Imaging Sciences	8 × 16	120	10, 18.5	29, 47	–	–	–	–	–	–	–	–	Morant et al21
SCANORA® 3D	Soredex (Helsinki, Finland)	7.5 × 10	85	30	46	42	50	30	–	45	148	1285	178	Pauwels et al30
Galileos Comfort Plus	Sirona Dental Systems	8.5 × 15	98	8, 10, 12 (20, 25, 30)	27, 34, 41 (84, 103, 122)	41	127	34	13	164	103	869	118	Current study
Alphard VEGA	Asahi Roentgen	10 × 10	80	68, 136	69, 146	196	907	308	17	1019	403	4544	659	Kim et al36
CB Mercuray	Hitachi (Tokyo, Japan)	10 × 10	80, 100, 120	100, 100, 150	60, 97, 145	–	–	–	–	–	–	–	–	Jadu et al38

H, height; kVp, kilovoltage peak; mAs, milliampere per second; W, width.

Bold values represent parameters used to produce standard or default scans.

^aAnterior.

^bPosterior.

^cUpgraded unit with additional filtration.

Table 6.

Adult phantom effective doses for standard or default exposures for small field of view (FOV) CBCT units (<10 cm height) views including mandible

Unit name	Manufacturer	FOV size H × W (cm)	kVp	mAs	Effective dose (µSv)	Bone marrow (µSv)	Bone surface (µSv)	Skin (µSv)	Oesophagus (µSv)	Brain (µSv)	Thyroid (µSv)	Salivary glands (µSv)	Remainder (µSv)	Study
3D Accuitomo 170	J Morita (Osaka, Japan)	4 × 4a	90	87.5	43	37	37	32	–	37	195	2120	70	Pauwels et al30
Veravieweposcs 3D	J Morita	4 × 4a	80	47	22	36	108	15	–	–	50	900	–	Al-Okshi et al43
CS 9000	Carestream Dental (Atlanta, GA)	4 × 5b	70	107	22	21	85	42	4	20	40	633	80	Current study
CS 9000	Carestream Dental	4 × 5a	70	107	40	78	35	24	–	290	251	709	86	Pauwels et al30
CS 9000	Carestream Dental	4 × 5a	70	107	38	37	150	10	10	20	110	1037	141	Current study
3D eXam®	Kavo (Bieberach, Germany)	4 × 16	120	18.5, 37	38, 76	–	–	–	–	–	–	–	–	Schilling and Geibel32
Alphard VEGA	Asahi Roentgen (Kyoto, Japan)	5 × 5	80	102, 153	62, 94	95	442	131	17	75	316	3570	507	Kim et al36
CS 9300	Carestream Dental	5 × 5b	84	60, 100	48, 81	56	207	44	37	32	424	2350	243	Current study
CS 9300	Carestream Dental	5 × 5a	84	60, 100	66, 127	87	332	54	66	51	384	3220	382	Current study
OP300 Maxio	Instrumentarium (Helsinki, Finland)	5 × 5b	90	11.7, 14.7, 18.7, 23.4	16, 20, 26, 32	14	52	16	15	8	162	621	83	Current study
Orthophos XG	Sirona Dental Systems (Bensheim, Germany)	5 × 5.5b	85	36, 51, 66 (72, 86, 101)	22, 31, 40 (49, 65, 84)	24	101	17	16	16	160	850	92	Current study
Orthophos XG	Sirona Dental Systems	5 × 5.5a	85	36, 51, 66 (72, 86, 101)	27, 38, 50, (61, 72, 83)	33	142	19	21	19	207	951	116	Current study
ProMax 3Dc	Planmeca Oy (Helsinki, Finland)	5 × 8	84	192	171	95	329	38	60	20	855	4048	659	Qu et al19
3D Accuitomo 170	J Morita	5 × 10	90	45, 79, 87.5, 154	87, 153, 169, 297	86	294	41	112	43	1216	4206	497	Current study
CS 9300	Carestream Dental	5 × 10	90	25	75	58	206	30	38	32	386	1910	248	Current study
3D Accuitomo 170	J Morita	5 × 14	90	45, 79, 87.5, 154	135, 237, 262, 461	258	810	64	127	74	1226	6363	865	Current study
3D Accuitomo 170	J Morita	5 × 17	90	45, 79, 87.5, 154	121, 212, 235, 414	193	618	62	152	65	1374	5539	734	Current study
Pan eXam Plus 3D	Kavo	6 × 4	90	23, 49	49, 115	–	–	–	–	–	–	–	–	Schilling and Geibel32
3D Accuitomo 170	J Morita	6 × 6	90	27, 45, 52.5, 79, 87.5, 154	37, 61, 72, 108, 120, 210	111	347	36	54	36	650	2814	381	Current study
3D Accuitomo 170	J Morita	6 × 6	90	27, 45, 52.5, 79, 87.5, 154	48, 80, 93, 148, 158, 252	156	514	44	75	55	745	3150	572	Current study
NewTom VGi	QR (Verona, Italy)	6 × 6	110	42.1	191	147	381	52	91	258	1150	4570	595	Current study
NewTom VGi	QR	6 × 6	110	29	130	103	268	72	58	187	667	3253	426	Current study
OP300 Maxio	Instrumentarium	6 × 8	90	11.7, 14.7, 18.7, 23.4	43, 54, 68, 86	86	282	28	33	18	305	1523	221	Current study
Pan eXam Plus 3D	Kavo	6 × 8	90	47, 79	110, 184	–	–	–	–	–	–	–	–	Schilling and Geibel32
iCAT® FLX	Imaging Sciences (Hatfield, PA)	6 × 16	90/120	6/10, 18.5, 37	8/34, 61, 127	61	124	18	33	51	290	1567	198	Ludlow and Walker22
iCAT NG	Imaging Sciences	6 × 16	120	18.5	45	33	33	25	–	46	251	973	172	Pauwels et al30
iCAT NG	Imaging Sciences	6 × 16	120	18.5, 37	58, 113	–	–	–	–	–	–	–	–	Davies et al34
iCAT NG	Imaging Sciences	6 × 16	120	10, 18.5	24, 39	–	–	–	–	–	–	–	–	Morant et al21
iCAT NG	Imaging Sciences	6 × 16	120	18.5, 37	75, 149	–	–	–	–	–	–	–	–	Roberts et al35
iCAT NG	Imaging Sciences	6 × 16	120	18.5	49	–	–	–	–	–	–	–	–	Theodorakou et al41
Picasso Trio	VATECH (Seoul, Korea)	7 × 12	85	91	81	62	57	56	–	39	583	1837	254	Pauwels et al30
Picasso Trio	VATECH	7 × 12	85	91, 127, (109)d	81, 123, (102)d	94	107	85	–	87	567	2410	342	Pauwels et al30
Picasso Trio	VATECH	7 × 12	85	127	123	126	156	113	–	134	551	2982	432	Pauwels et al30
DCT PRO	VATECH	7 × 16	90	105	180	76	240	20	150	20	2360	2280	377	Qu et al40
SCANORA® 3D	Soredex (Helsinki, Finland)	7.5 × 10	85	30	47	34	35	29	–	25	352	1052	147	Pauwels et al30
SCANORA 3D	Soredex	7.5 × 10	85	30	45	37	39	31	–	31	240	1117	155	Pauwels et al30
3D Accuitomo 170	J Morita	8 × 8	90	45, 79, 87.5, 154	92, 162, 180, 316	180	560	103	85	83	921	4257	564	Current study
3D eXam	Kavo	8 × 8	120	18.5, 37	62, 122	–	–	–	–	–	–	–	–	Schilling and Geibel32
CS 9300	Carestream Dental	8 × 8	90	32	75	59	219	33	32	69	330	1935	254	Current study
iCAT FLX	Imaging Sciences	8 × 8	90/120	6/10, 18.5, 37	5/23, 44, 85	85	139	23	42	58	222	1172	149	Ludlow and Walker22
iCAT NG	Imaging Sciences	8 × 8	120	10, 18.5	18, 29	–	–	–	–	–	–	–	–	Morant et al21
NewTom VGi	QR	8 × 8	110	6.3 (38.7)	61 (206)	53	139	43	26	174	258	1494	208	Current study
OP300 Maxio	Instrumentarium	8 × 8	90	11.7, 14.7, 18.7, 23.4	49, 61, 78, 97	93	308	35	35	24	335	1754	255	Current study
Orthophos XG	Sirona Dental Systems	8 × 8	85	36, 51, 66 (72, 86, 101)	48, 67, 91 (117, 144, 166)	53	240	28	32	53	298	1718	226	Current study
Prexion 3D high res	TeraRecon (Foster City, CA)	8 × 8	90	148	388	325	1508	264	133	783	1800	9372	1309	Ludlow and Ivanovic15
Prexion 3D standard	TeraRecon	8 × 8	90	76	189	164	760	135	53	383	683	4761	684	Ludlow and Ivanovic15
ProMax 3D	Planmeca Oy	8 × 8	84	19.6, 169	28, 122	88	121	145	–	53	1021	2576	346	Pauwels et al30
ProMax 3D	Planmeca Oy	8 × 8	84	19.6	18	–	–	–	–	–	–	–	–	Theodorakou et al41
Promax 3D	Planmeca Oy	8 × 8	84	72, 96	488, 652	468	2170	339	120	600	1267	12 939	1846	Ludlow and Ivanovic15
ProMax 3Dc	Planmeca Oy	8 × 8	84	22.4, 96, 120, 144, 168, 192, 192	30, 102, 169, 216, 272, 298, 306	255	883	163	94	215	1101	6582	962	Qu et al19
Veravieweposcs 3D	J Morita	8 × 8	70	51	73	55	57	69	–	40	330	1956	85	Pauwels et al30
NewTom VGi	QR	8 × 12	110	6.1, (39.1)	82 (280)	87	219	49	41	237	316	1942	275	Current study
Kodak 9500	Carestream Dental	8 × 15	90	108	92	85	84	51	–	91	541	2166	304	Pauwels et al30
OP300 Maxio	Instrumentarium	8 × 15	90	22.5, 28.4, 36, 45	76, 96, 121, 152	127	431	71	66	58	550	2680	409	Current study
3D eXam	Kavo	8 × 16	120	10, 37 (18.5)	45, 170 (88)	–	–	–	–	–	–	–	–	Schilling and Geibel32
iCAT FLX	Imaging Sciences	8 × 16	90/120	6/10, 18.5, 37	8/39, 70, 148	70	150	23	40	95	329	1693	225	Ludlow and Walker22
iCAT NG	Imaging Sciences	8 × 16	120	18.5, 37	65, 134	76	151	33	17	150	183	1639	235	Grunheid et al33
Galileos Comfort plus	Sirona Dental Systems	8.5 × 15	98	8, 10, 12 (20, 25, 30)	29, 37, 45, (92, 113, 133)	40	118	21	26	23	225	894	121	Current study
CS 9500	Carestream Dental	9 × 15	80, 85, 90	86.4, 108, 108	76, 98, 166	113	386	55	54	264	533	1680	313	Ludlow10
Alphard VEGA	Asahi Roentgen	10 × 10	80	68, 136	85, 184	217	1009	147	43	496	872	6233	788	Kim et al36
CB Mercuray	Hitachi (Tokyo, Japan)	10 × 10	120	150	407	466	2161	344	110	2950	1300	9006	1355	Ludlow and Ivanovic15
CB Mercuray	Hitachi	10 × 10	80, 100, 120	100, 100, 150	148, 261, 421	–	–	–	–	–	–	–	–	Jadu et al38

H, height; kVp, kilovoltage peak; mAs, milliampere per second; W, width.

Bold values represent parameters used to produce standard or default scans.

^aPosterior.

^bAnterior.

^cUpgraded unit with additional filtration.

^dAverage of high- and low-dose protocols.

Table 7.

Adult phantom equivalent and effective doses for standard or default exposures for small field of view (FOV) CBCT units (<10 cm height)—temporomandibular joint views

Unit name	Manufacturer	FOV size H × W(cm)	kVp	mAs	Effective dose (µSv)	Bone marrow (µSv)	Bone surface (µSv)	Skin (µSv)	Oesophagus (µSv)	Brain (µSv)	Thyroid (µSv)	Salivary glands (µSv)	Remainder (µSv)	Study
CS 9000	Carestream Dental (Atlanta, GA)	4 × 5	68, 70, 70	68, 86, 108	10, 14, 21	18	18	–	–	218	23	231	43	Lukat et al42
NewTom VGi	QR (Verona, Italy)	8 × 8	110	22, 93	45, 129	40	102	62	–	150	20	2130	–	Al-Okshi et al43
NewTom VGi	QR	8 × 12	110	19.1	56	48	122	74	–	230	20	2400	–	Al-Okshi et al43
CB Mercuray	Hitachi (Tokyo, Japan)	10 × 10	120	150	279	244	1882	308	53	6500	538	4810	190	Librizzi et al37

H, height; kVp, kilovoltage peak; mAs, milliampere per second; W, width.

Bold values represent parameters used to produce standard or default scans.

Table 8.

Child phantom equivalent and effective doses for standard or default exposures for medium and large field of view (FOV) CBCT units (>10 cm height)

Unit name	Manufacturer	FOV size H × W(cm)	kVp	mAs	Effective dose (µSv)	Bone marrow (µSv)	Bone surface (µSv)	Skin (µSv)	Oesophagus (µSv)	Brain (µSv)	Thyroid (µSv)	Salivary glands (µSv)	Remainder (µSv)	Study
ProMax mid-stiched	Planmeca Oy (Helsinki, Finland)	16 × 16	90	(72.3, 90) 217, 244 “289, 362”	(88, 107) 277, 307 “362, 488”	318	1118	222	112	2829	2154	3706	145	Current study
Iluma	Imtec (Ardmore, OK)	19 × 19	120	20	46	–	–	–	–	–	–	–	–	Vassileva and Stoyanov16
Newtom 3G	Cefla Dental Group (Imola, Italy)	20 × 20	110	AEC	56	71	182	22	44	430	595	531	85	Current study
3D Accuitomo 170	J Morita (Osaka, Japan)	10 × 10	90	45, 79, 87.5, 154	160, 281, 311, 548	151	580	354	99	1141	2804	5736	825	Current study
CS 9300	Carestream Dental (Atlanta, GA)	10 × 10	80	25	86	43	187	47	30	171	846	1489	223	Current study
3D Accuitomo 170	J Morita	10 × 14	90	87.5	237	–	–	–	–	–	–	–	–	Theodorakou et al41
3D Accuitomo 170	J Morita	10 × 14	90	45, 79, 87.5, 154	183, 321, 355, 626	209	804	319	114	1693	2993	6440	947	Current study
Newtom VG	Cefla Dental Group	11 × 15	110	Auto	114	–	–	–	–	–	–	–	–	Theodorakou et al41
iCAT® FLX	Imaging Sciences (Hatfield, PA)	11 × 16	90/120	6/10, 18.5	13, 56, 115	115	190	81	53	391	1001	2045	302	Ludlow and Walker22
CS 9300	Carestream Dental	11 × 17	80	25.6, 41.2	110, 178	120	524	111	81	850	1648	2751	425	Current study
3D Accuitomo 170	J Morita	12 × 17	90	87.5	282	–	–	–	–	–	–	–	–	Theodorakou et al41
3D Accuitomo 170	J Morita	12 × 17	90	45, 79, 87.5, 154	212, 353, 430, 769	244	940	307	253	2039	4265	6622	1004	Current study
OP300 Maxio	Instrumentarium (Helsinki, Finland)	13 × 15	90	29, 36, 45, 72	93, 108, 134, 215	58	225	66	38	407	765	1668	254	Current study
iCAT FLX	Imaging Sciences	13 × 16	90/120	6/10, 18.5	18, 70, 120	120	211	82	53	731	1003	2038	303	Ludlow and Walker22
iCAT NG	Imaging Sciences	13 × 16	120	18.5	134	–	–	–	–	–	–	–	–	Theodorakou et al41
CS 9300	Carestream Dental	13.5 × 17	80	45.2	189	118	509	155	78	964	1741	3075	458	Current study
Galileos Comfort plus	Sirona Dental Systems (Bensheim, Germany)	15 × 15	98	6, 8 (15, 20)	39, 52, (122, 160)	26	91	25	22	185	384	589	91	Current study
Newtom 3G	Cefla Dental Group	15 × 15	110	Auto	94	100	265	56	44	679	875	1250	190	Current study

H, height; kVp, kilovoltage peak; mAs, milliampere per second; W, width.

Bold values represent parameters used to produce standard or default scans.

Table 9.

Child phantom equivalent and effective doses for standard or default exposures for small field of view (FOV) CBCT units (<10 cm height) maxillary views

Unit name	Manufacturer	FOV size H × W(cm)	kVp	mAs	Effective dose (µSv)	Bone marrow (µSv)	Bone surface (µSv)	Skin (µSv)	Oesophagus (µSv)	Brain (µSv)	Thyroid (µSv)	Salivary glands (µSv)	Remainder (µSv)	Study
3D Accuitomo 170	J Morita (Osaka, Japan)	4 × 4a	90	87.5	28	–	–	–	–	–	–	–	–	Theodorakou et al41
CS 9000	Carestream Dental (Atlanta, GA)	4 × 5a	70	85.6	16	–	–	–	–	–	–	–	–	Theodorakou et al41
CS 9300	Carestream Dental	5 × 5a	75	48, 80	41, 63	26	118	36	12	70	227	1930	243	Current study
CS 9300	Carestream Dental	5  ×  5b	75	48, 80	47, 79	38	172	33	19	62	384	2165	282	Current study
OP300 Maxio	Instrumentarium (Helsinki, Finland)	5 × 5a	90	11.7, 14.7, 18.7, 23.4	16, 20, 26, 32	6	21	20	3	44	53	438	67	Current study
3D Accuitomo 170	J Morita	5 × 10	90	45, 79, 87.5, 154	58, 102, 113, 198	44	163	228	26	355	423	2988	433	Current study
CS 9300	Carestream Dental	5 × 10	80	25	45	18	80	27	9	78	161	1269	179	Current study
3D Accuitomo 170	J Morita	5 × 14	90	45, 79, 87.5, 154	85, 149, 165, 290	65	244	245	37	564	568	4423	648	Current study
3D Accuitomo 170	J Morita	5 × 17	90	45, 79, 87.5, 154	91, 160, 177, 312	77	289	177	39	508	575	4974	701	Current study
3D Accuitomo 170	J Morita	6  × 6b	90	27, 45, 52.5, 79, 87.5, 154	32, 53, 61, 92, 102, 180	37	138	201	21	531	343	2580	406	Current study
OP300 Maxio	Instrumentarium	6 × 8	90	11.7, 14.7, 18.7, 23.4	27, 34, 43, 54	9	34	39	5	107	90	708	108	Current study
iCAT® FLX	Imaging Sciences (Hatfield, PA)	6 × 16	90/120	6/10, 18.5	5, 23, 39	39	43	69	11	252	158	889	142	Ludlow and Walker22
iCAT NG	Imaging Sciences	6 × 16	120	18.5	43	–	–	–	–	–	–	–	–	Theodorakou et al41
Galileos Comfort plus	Sirona Dental Systems (Bensheim, Germany)	8.5 × 15	98	6, 8 (15, 20)	21, 32 (72, 98)	12	41	23	6	175	92	496	72	Current study

H, height; kVp, kilovoltage peak; mAs, milliampere per second; W, width.

Bold values represent parameters used to produce standard or default scans.

^aAnterior.

^bPosterior.

Table 10.

Child phantom equivalent and effective doses for standard or default exposures for small field of view (FOV) CBCT units (<10 cm height) views including mandible

Unit name	Manufacturer	FOV size H × W(cm)	kVp	mAs	Effective dose (µSv)	Bone marrow (µSv)	Bone surface (µSv)	Skin (µSv)	Oesophagus (µSv)	Brain (µSv)	Thyroid (µSv)	Salivary glands (µSv)	Remainder (µSv)	Study
CS 9300	Carestream Dental	5 × 5a	75	48, 80	44, 64	29	135	29	18	44	332	1776	223	Current study
CS 9300	Carestream Dental	5 × 5b	75	48, 80	56, 86	41	187	36	23	57	532	2144	285	Current study
OP300 Maxio	Instrumentarium (Helsinki, Finland)	5 × 5a	90	11.7, 14.7, 18.7, 23.4	33, 42, 53, 67	22	86	3	10	17	328	641	82	Current study
3D Accuitomo 170	J Morita (Osaka, Japan)	5 × 10	90	45, 79, 87.5, 154	152, 266, 295, 519	151	591	60	107	127	3382	5023	660	Current study
CS 9300	Carestream Dental	5 × 10	80	25	61	30	134	24	20	46	457	1365	192	Current study
3D Accuitomo 170	J Morita	5 × 14	90	87.5	214	–	–	–	–	–	–	–	–	Theodorakou et al41
3D Accuitomo 170	J Morita	5 × 14	90	45, 79, 87.5, 154	152, 267, 296, 521	187	739	70	128	142	3356	3356	645	Current study
3D Accuitomo 170	J Morita	5 × 17	90	45, 79, 87.5, 154	170, 299, 331, 582	196	773	92	113	186	3326	5937	833	Current study
3D Accuitomo 170	J Morita	6 × 6a	90	27, 45, 52.5, 79, 87.5, 154	46, 77, 90, 136, 150, 265	67	259	49	42	137	1031	3873	469	Current study
3D Accuitomo 170	J Morita	6  ×  6b	90	27, 45, 52.5, 79, 87.5, 154	71, 118, 141, 208, 227, 375	108	419	68	72	202	1549	5558	725	Current study
OP300 Maxio	Instrumentarium	6 × 8	90	11.7, 14.7, 18.7, 23.4	61, 77, 97, 122	34	135	6	17	37	576	1153	164	Current study
iCAT® FLX	Imaging Sciences (Hatfield, PA)	6 × 16	90/120	6/10, 18.5	9, 43, 73	73	98	34	30	106	530	1654	224	Ludlow and Walker22
iCAT NG	Imaging Sciences	6 × 16	120	18.5	63	–	–	–	–	–	–	–	–	Theodorakou et al41
3D Accuitomo 170	J Morita	8 × 8	90	45, 79, 87.5, 154	128, 225, 249, 439	117	454	100	74	324	2269	5023	687	Current study
CS 9300	Carestream Dental	8 × 8	80	32	82	42	183	30	25	80	780	1529	219	Current study
iCAT FLX	Imaging Sciences	8 × 8	90/120	6/10, 18.5	7, 34, 60	60	77	20	20	134	403	1401	191	Ludlow and Walker22
OP300 Maxio	Instrumentarium	8 × 8	90	11.7, 14.7, 18.7, 23.4	77, 97, 123, 153	43	167	37	23	186	685	1416	210	Current study
ProMax 3D	Planmeca Oy (Helsinki, Finland)	8 × 8	84	19.6	24	–	–	–	–	–	–	–	–	Theodorakou et al41
iCAT FLX	Imaging Sciences	8 × 16	90/120	6/10, 18.5	12, 50, 85	85	126	39	35	192	659	1754	248	Ludlow and Walker22
Galileos® ComfortPLUS	Sirona Dental Systems (Bensheim, Germany)	8.5 × 15	98	6, 8 (15, 20)	29, 44, (99, 136)	20	71	11	20	17	327	404	61	Current study

H, height; kVp, kilovoltage peak; mAs, milliampere per second; W, width.

Bold values represent parameters used to produce standard or default scans.

^aAnterior.

^bPosterior.

Tables 11–14 provide summary data for adult and child doses and include mean values with standard deviations for standard or default exposure equivalent and effective doses. Lower numbers of units with equivalent doses reflect the practice of some studies that have included fewer weighted tissues in their calculation of effective dose or report only effective dose. Standard adult exposure settings resulting in average adult effective doses of 212 µSv for large FOVs, 177 µSv for medium FOVs and 84 µSv for small FOVs are found in Table 11. Small adult FOVs producing average effective doses of 53 µSv for maxillary views and 102 µSv for mandibular views are seen in Table 12. Because data for few large FOVs were found for child phantoms, large and medium FOVs were combined in Table 13. The average effective dose for large or medium FOVs was 175 µSv. The average child effective dose for small FOVs was 103 µSv. When child phantom small FOVs were analysed by arch location, an average dose of 67 µSv was seen for maxillary views and of 128 µSv was seen for mandibular views, as seen in Table 14.

Table 11.

Average equivalent and effective doses (μSv) for an adult using standard exposure settings of dental CBCT units

Field of view size	Bone marrow	Bone surface	Skin	Oesophagus	Brain	Thyroid	Salivary glands	Remainder	Effective dose
Large
Units reported	16	16	16	14	16	16	16	16	23
Mean	359	1457	189	122	2182	1130	3484	475	212
SD	428	2081	215	125	2601	1616	3465	502	212
Medium
Units reported	32	32	32	26	32	32	32	32	43
Mean	233	844	163	75	1211	762	3675	442	177
SD	240	1194	153	53	1911	657	2643	325	137
Small
Units reported	77	79	78	57	78	77	79	74	101
Mean	94	299	62	43	211	413	2259	316	84
SD	81	376	61	38	746	433	2027	295	78

SD, standard deviation.

Table 14.

Small field of view (FOV)—average equivalent and effective doses (μSv) by arch for a 10-year-old child using standard exposure settings of dental CBCT units

FOV location	Bone marrow	Bone surface	Skin	Oesophagus	Brain	Thyroid	Salivary glands	Remainder	Effective dose
Maxilla
Units reported	11	11	11	11	11	11	11	11	14
Mean	34	122	100	17	250	279	2078	298	67
SD	23	89	92	13	205	190	1553	223	54
Mandible
Units reported	17	17	17	17	17	17	17	17	20
Mean	77	273	42	46	120	1207	2589	360	128
SD	57	234	28	38	82	1138	1813	251	101

SD, standard deviation.

Table 12.

Small field of view (FOV)—average equivalent and effective doses (μSv) for an adult using standard exposure settings of dental CBCT units

FOV location	Bone marrow	Bone surface	Skin	Oesophagus	Brain	Thyroid	Salivary glands	Remainder	Effective dose
Maxilla
Units reported	25	27	27	20	27	25	27	25	38
Mean	85	236	53	18	153	155	1374	213	53
SD	75	215	40	13	253	104	949	155	38
Mandible
Units reported	48	48	48	36	47	48	48	47	59
Mean	100	316	61	57	112	569	2745	380	102
SD	84	388	63	41	151	477	2324	337	88

SD, standard deviation.

Table 13.

Average equivalent and effective doses (μSv) by field of view (FOV) size for a 10-year-old child using standard exposure settings of dental CBCT units

FOV size	Bone marrow	Bone surface	Skin	Oesophagus	Brain	Thyroid	Salivary glands	Remainder	Effective dose
Large or medium
Units reported	13	13	13	13	13	13	13	13	18
Mean	130	448	142	79	962	1621	2918	404	175
SD	83	331	119	61	791	1148	2117	319	115
Small
Units reported	28	28	28	28	28	28	28	28	34
Mean	60	213	65	34	171	843	2388	336	103
SD	50	202	67	33	154	997	1705	238	89

SD, standard deviation.

Table 15 provides p-values for an ANOVA of effective dose and equivalent doses for each of the weighted tissues that are typically included in head and neck dosimetry studies. The ANOVA model investigated the effects of FOV and phantom type. Tukey honest significant difference results are provided for FOV for statistically significant factors. With the exception of remainder tissues, all weighted tissues and effective dose demonstrated significantly increased dose with increased FOV size. With the exception of thyroid dose, which was significantly greater in child exposures, no differences were seen in equivalent or effective dose owing to phantom type. A separate analysis investigated dose differences related to maxillary or mandibular location for small FOVs for child and adult phantoms. Significantly higher doses were associated with mandibular field positions for the oesophagus, thyroid, salivary gland and remainder tissues as well as effective dose. Once again, only the thyroid tissue demonstrated significantly higher doses in child phantoms than in adult phantoms.

Table 15.

ANOVA: all volume data models include phantom type and field of view (FOV); small volume data model includes phantom type and arch

p-values	Post hoc statistical test	Model/HSD variable level	Bone marrow	Bone surface	Skin	Oesophagus	Brain	Thyroid	Salivary glands	Remainder	Effective dose
		Phantom	0.0538	0.1788	0.8268	0.4821	0.6351	0.0005	0.7143	0.8926	0.6198
		FOV size	<0.0001	<0.0001	<0.0001	<0.0001	<0.0001	0.0004	0.008	0.0868	<0.0001
All FOVs	Tukey HSD Test	Large	A	A	A	A	A	A	A	A	A
		Medium	B	B	A	A	B	A	A B	A	A
		Small	C	C	B	B	C	B	B	B	B
Small FOV		Phantom	0.0406	0.2986	0.6479	0.3281	0.3009	0.0009	0.6789	0.7445	0.1706
		FOV location	0.1262	0.1104	0.384	<0.0001	0.0779	<0.0001	0.0042	0.0198	0.0002

HSD, honestly significant difference.

Bold values indicated statistical significance at α < 0.05.

FOV levels not connected by the same letter (A, B, C) are significantly different.

Distribution of mean effective dose components from large FOV CBCT imaging of adult phantoms is displayed graphically in Figure 1. A similar graphic for combined large or medium FOVs is provided for child phantoms in Figure 2. The graphics demonstrate the greater contribution of thyroid exposure to effective dose in the child (37%) than in adult (20%).

Figure 1.

Distribution of effective dose components in an adult phantom for large field of view dental CBCT.

Figure 2.

Distribution of effective dose components from large or medium field of view CBCT imaging of a child phantom.

Table 16 provides summary statistics for volume DAP and DHP and the ratios of effective dose to each of these values. The coefficient of variation increases with reduction in FOV area or height. The coefficient of variation is consistently lower for E/DHP than for E/DAP ratios regardless of the FOV size. This suggests that FOV height may be a better predictor of effective dose than FOV area.

Table 16.

Dose–area product (DAP) and dose–volume height (DHP) calculations and derived conversion coefficients for calculating effective dose derived from phantom data

Phantom	FOV	Descriptive statistic	DAP (mGy cm2)	DHP (mGy cm)	Conversion coefficient
					E/DAP (µSv mGy−1 cm−2)	E/DHP (µSv mGy−1 cm−1)
Adult	Large	Mean	1229	64	0.17	3.44
		SD	1240	66	0.04	0.66
		c.v.	–	–	23%	19%
	Medium	Mean	683	46	0.27	3.90
		SD	574	36	0.08	0.78
		c.v.	–	–	28%	20%
	Small maxilla	Mean	61	7	1.28	8.29
		SD	55	5	0.62	2.66
		c.v.	–	–	48%	32%
	Small mandible	Mean	175	18	0.87	6.63
		SD	161	18	0.51	1.87
		c.v.	–	–	59%	28%
Child	Large and medium	Mean	529	35	0.35	5.07
		SD	362	23	0.10	0.53
		c.v.	–	–	29%	11%
	Small maxilla	Mean	121	11	0.85	6.87
		SD	129	8	0.42	0.74
		c.v.	–	–	49%	11%
	Small mandible	Mean	153	16	1.00	8.47
		SD	126	11	0.43	2.98
		c.v.	–	–	44%	35%

c.v., coefficient of variation; E, effective dose; FOV, field of view; SD, standard deviation.

Analysis by phantom and field of view size.

Table 17 displays average absolute error between phantom-based calculations of the effective dose and E/DAP or E/DHP ratios as coefficients for effective dose estimation. The magnitude of error increases as the FOV size is reduced. The average absolute error for all volumes was 35% for E/DAP and decreased to 19% for E/DHP. This difference was statistically significant (p < 0.0001).

Table 17.

Absolute error between phantom effective dose and dose calculated with dose–area product (DAP) and dose–height product (DHP) conversion coefficients

Field of view	Phantom	n	Ephantom − EDAP (µSv)	Ephantom − EDHP (µSv)	Probability < t
Large	Child	–	–	–	–
	Adult	16	36.4	28.8	0.0372
Medium	Child	13	32.5	15.2	0.0183
	Adult	32	39.1	27.5	0.0611
Small maxilla	Child	11	41.3	5.1	0.0213
	Adult	31	48.2	23.0	0.0050
Small mandible	Child	17	45.6	38.7	0.2851
	Adult	48	56.1	25.9	<0.0001
All		168	45.7	25.1	<0.0001
% diff from E		–	35%	19%	–

E, effective dose.

Bold values indicated statistical significance at α < 0.05.

Discussion

Reported dosimetry for standard CBCT exposure settings demonstrated significant reductions in effective dose associated with the use of small FOV sizes. While a trend of dose reduction from large to medium FOVs was seen, this was not statistically significant. The absence of a significant dose–FOV relationship is likely related to two factors. Increasing FOV extends anatomic coverage superiorly increasing the amounts of brain and bone coverage with little increase in exposure of other weighted tissues. The resulting proportional increase in effective dose is smaller than that seen when small fields centred on the dentoalveolar area expand both cranially and caudally in medium FOV volumes. The wide range of effective doses produced using standard settings by different CBCT units is another factor affecting the statistical significance of differences that may be present between medium and large FOVs. Standard deviations associated with dose values for these volumes were on the same order as calculated means indicating substantial variability among devices. Although not assessed in detail in this study, exposure variability also increases substantially when one includes the range of protocol options offered by many manufacturers. Doses for the same FOV may have as much as a 15-fold difference between low-dose and high-resolution protocols.²²

This study has focused on standard or default exposures. These are protocols recommended by the manufacturer of the CBCT unit for imaging of average or typical patients. It should be noted that manufacturers may change exposures associated with the standard designation over time. Examples of this are seen with Iluma and CB Mercuray devices. Initially, the highest exposure protocols were recommended for these devices. Later, much lower exposure protocols became the recommended “standard”. This is reflected in study data from different investigators in Table 3. While marketing materials were changed to reflect updated standard protocols, the effect on selection of imaging parameters by end users is unclear. Differences between recommended use of products and actual clinical application are not new to dental radiology. Continued use of round cones and D-speed film by many dental practitioners in the face of many years of recommendations by the American Dental Association and the National Commission on Radiation Protection and Measurements encouraging the use of rectangular collimation and high-speed receptors is a prominent example.^44,45

Methodologic errors in dosimetry and effective dose calculation

Regardless of the FOV size, remainder tissues accounted for <10% of the effective dose calculations using ICRP 1990¹² tissue weights.¹⁸ The brain was the only tissue contributing significant dose in the remainder group. Application of ICRP 2007¹³ tissue weights resulted in remainder doses contributing 27–42% to effective dose depending on FOV size and location.¹⁸ Studies excluding remainder tissues from their dose calculations significantly underestimate dose, and a number of these were excluded from this report.^46,47 A citation that did not include oral mucosa, which was added to the remainder group in the 2007 ICRP calculation of effective dose was also excluded.⁴⁸ The oral mucosa is directly exposed in any maxillofacial CBCT scan and as a component of the remainder group has a tissue weight of 0.0092. This is nearly the tissue weight of the salivary glands (0.01). Similarly, the extrathoracic region is exposed in most dental CBCT scans. Together, the contribution of these tissues to effective dose exceeds that of the oesophagus, skin, bone surface and brain combined and cannot be overlooked in a calculation of stochastic risk.¹⁸

Sampling is an important component of dosimetry, and the sampling strategy is critical to both internal validity of a dosimetry study and its extensibility to other studies or patient populations. One approach that has been taken is to sample doses over a regularly spaced grid throughout a phantom. This approach requires many dosemeters and results in time consuming and expensive study protocols. An additional complication is that the tissues and organs of particular interest for radiation biology are not uniformly distributed. A uniform grid of dosemeters may not coincide with the location of a tissue of interest. An alternate approach is to place dosemeters only in the weighted tissues used in the calculation of the effective dose. A uniform distribution of dosemeters within the selected tissue is still a resource intensive choice, so efforts to strategically locate dosemeters within a tissue such that the average dose of a limited number of dosemeters reasonably reflects that of a uniform distribution of dosemeters is desirable. We have used this approach in measuring calvarial bone and marrow doses, where a limited number of strategically positioned dosemeters are used to reflect dose to the entire skull. For tissues that are incompletely contained in the maxillofacial area, sampling of the directly exposed portion of the tissue multiplied by the percentage of total tissue that is directly exposed provides a reasonable estimation of organ dose. This approach is taken with skin, muscle, bone, lymph nodes and oesophagus in our studies.^{10,15,17,22,49} Studies that fail to account for body-wide distributions of tissue produce overestimations of equivalent doses for the bone, bone marrow, oesophagus, lymphoid tissue, muscle and skin.^50,51 This error may result in a 10- to 20-fold overestimation of specific organ dose with a concomitant exaggeration of effective dose. The bone marrow, which varies in quantity and distribution by patient age is accorded different percentages for child and adult phantoms.²² The same dosemeters may be used to calculate bone and bone marrow dose; however, it is important to account for differences in X-ray attenuation efficiency when calculating absorbed doses. The higher effective atomic number of bone leads to increased photoelectric interactions and increased dose to this tissue. Mass attenuation coefficients for bone and soft tissue are available from a variety of sources and are usually calculated using mono-energetic photons beams.²⁹ Bone/muscle attenuation ratios can be calculated from these data. A linear fit of ratios from 40 to 80 kV is adequate to cover the mean beam energy range produced by CBCT units. In our studies, the effective beam energy of the highly filtered, low ripple, polychromatic beams used for CBCT is estimated to be two-thirds of the peak beam energy. Using this assumption and an equation developed from the linear fit of monochromatic data, bone muscle attenuation ratios from 1.97 for a peak kilovoltage of 120 kVp to 3.63 for 80 kVp are used. Studies that fail to adjust for attenuation differences of the bone and soft tissues may underestimate bone doses by a factor of 2–4×.^30,42 Examining the ratio of the bone to bone marrow doses using the means in Table 11 suggests a range of 3.1–4.1× average underestimation of the bone dose if dosemeter values are not corrected for bone attenuation efficiency. This will lead to a 2–5% underestimation of effective dose depending on the size of the FOV.

Tissues that are completely outside the field of direct exposure are not sampled in our dosimetry protocol. These tissues account for 75% of the weighted tissues in a full body exposure; however, their indirect exposure in maxillofacial examinations accounts for <2% of effective dose.^18,51,52 Although the oesophagus is typically outside the field of direct exposure during CBCT scans, it is potentially exposed to scatter radiation. Because the oesophagus surrounds an open air space, caudally directed scatter photons can extend into the upper oesophageal passage exposing the mucosal walls. For this reason, an organ fraction of 10% is used in calculating oesophageal dose. Because of the oesophagus tissue weight of 0.04 in the calculation of the effective dose, this organ is as or more important than skin dose. Figure 1 suggests that studies that fail to measure oesophageal dose underestimate effective dose by as much as 2%.³⁰

While the use of additional dosemeters in indirectly or directly exposed tissues may increase the precision of calculation of organ dose, it does not guarantee an increase in the accuracy of the calculation of effective dose. Patient positioning for CBCT can vary between devices and operators, and this can have a pronounced effect on dose. Duplication of phantom position when comparing dose differences owing to technique is critical if results are to reflect differences owing to protocol rather than confounding owing to differences in patient position. A 10° rotation in phantom position was noted to produce a 92% difference in dose to the thyroid.¹⁸ Similarly, while caudal–cranial positioning differences of the FOV by a few millimetres may be clinically acceptable, when it results in direct exposure of the thyroid area, it can lead to 3- to 4-fold increases in dose to the thyroid gland and increases in effective dose by as much as 30%.⁵³

The allure and disappointment of dose–area product

While the use of DAP has been advocated as a measure of dose for CBCT units,^54,55 its accuracy as a measure of risk is debatable. In relating DAP to effective dose, a conversion coefficient must be used. Using a single CBCT device, a recent study calculated conversion coefficients for DAP to effective dose for large to small FOVs and found that a 3.8-fold range of values (0.038–0.146 µSv mGy⁻¹ cm⁻²) was required for different field sizes and anatomic locations.³⁶ A 7.5-fold range of E/DAP values was calculated from the adult phantom data in this study encompassing a large number of measurements on CBCT devices of varying beam energies. A 2.9-fold range of E/DAP values was calculated for a more limited set of child phantom data in this study. The use of volume height in the place of projection area resulted in statistically improved accuracy in the estimation of effective dose but still led to a range in the conversion coefficient of 2.4-fold for adult imaging and a 1.7-fold range for child imaging. A best-case scenario of using conversion factors specific for FOV size, arch location and patient type still resulted in an average absolute error of 35% of the calculated effective dose when using DAP. Improvement in average absolute error to 19% was seen when volume height was substituted for beam area. This improvement is intriguing as beam height may be easily substituted for beam area in dose calculations. While dose to salivary glands was used in this study, dose measured in air or air kerma are typically used in DAP calculations. These result in different numerical values for effective dose conversion factors but could be hypothesized to provide similar variability for DAP and DHP calculations. While this warrants further investigation, it is apparent that development of universal conversion coefficients to translate simple measures of exposure to patient dose with the goal of risk estimation is problematic.

Why does dose–height product correlate more closely than dose–area product to effective dose?

The dimensions of the rectangular X-ray beam are used to calculate the area portion of DAP. For a homogenous object that is at least the size of the X-ray beam cross section, the product of exposure and beam area will provide a good correlation with absorbed dose. For maxillofacial imaging, the imaged structures are non-homogenous with respect to tissue density, the shape of tissues imaged and the distribution of tissues with respect to radiation sensitivity. Beams of wider width may more than cover the horizontal dimension of the face. The portion of the X-ray beam that extends beyond the face increases beam area but adds negligibly to patient dose because it exposes the patient to only a small amount of scatter photons from the beam interaction with air. The vertical dimensions of even large FOVs rarely exceed the dimension of the maxillofacial area. Therefore, an increase in height of the FOV almost always results in an increased volume of exposed tissue. But DHP is a better correlate with effective dose even for small FOVs. This may be explained by the vertical distribution of radiosensitive organs. Small amounts of direct exposure of an organ contribute more to dose than do larger amounts of scatter radiation. For instance, a small amount of direct exposure of the thyroid gland may result in a dramatic increase in both thyroid and effective dose.⁵³ Mandibular locations of small FOVs provide greater exposure of the thyroid and submandibular salivary glands than do same size maxillary views. Increasing height of the FOV brings new and potentially radiosensitive tissues into the area of direct exposure, while increasing width of the beam simply increases dose to tissues already being exposed.

Conclusions

Given a choice, dentists prefer images with technical factors that provide high signal-to-noise ratios and high resolution. Dentists requesting images from an imaging centre or providing examinations in their own offices may not understand the risk implications of using higher doses to obtain image volumes. If “pretty pictures” are being obtained when a “just diagnostic” image is needed, we are doing the patient a disservice.¹⁸ As imaging professionals, it is our responsibility to educate our colleagues in other specialities and general dentistry about the risk differences between “diagnostic” and “pretty”. This is reason enough for dosimetry research. Recognizing that diagnostic imaging is the single greatest source of exposure to ionizing radiation for the US population that is controllable, the National Commission on Radiation Protection and Measurements has introduced a modification of the as low as reasonably achievable concept. ALADA represents “as low as diagnostically acceptable”.⁵⁶ Implementation of this concept will require evidence-based judgments of the level of image quality required for specific diagnostic tasks, and exposures and doses associated with this level of quality. Little research is currently available in this area.

For a dosimetry study reporting effective dose to be comparable with the contemporary studies cited in this review, it should incorporate the following elements: the dose calculation should follow the ICRP 2007 recommendations including new and adjusted tissue weights from all weighted tissues in the head and neck area. The location of dosemeters or points of measurement should be specified. When an entire organ is not exposed, such as bone or skin surface, the strategy for extrapolating dose for the entire organ from incomplete sampling should be described. Adjustment for the mass attenuation differences of the bone and soft tissue should be made and described. Adjustments for sensitivity of the dosimetry system for the mean beam energy of the X-ray source should also be described.

Dosimetry research involves many variables in addition to the examination that is being assessed. Decisions affecting these variables may have a profound impact on accuracy, validity and extensibility of results. Currently, anthropomorphic phantom dosimetry, when properly executed, provides the most accurate estimation of effective dose. Large exposure ranges make CBCT doses difficult to generalize. The use of DAP with average conversion coefficients to calculate dose results in significant inaccuracy. The use of DHP as a metric for estimating effective dose improves accuracy and warrants further investigation.

Conflict of interest

Travel support and honoraria were provided to Dr Ludlow by AFP Imaging, Carestream Dental, Instrumentarium, and Sirona Dental Systems. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.