Effective dose and risk assessment from film tomography used for dental implant diagnostics

Transcription

1 Effective dose and risk assessment from film tomography used for dental implant diagnostics N.L. Frederiksen, B.W. Benson and T.W. Sokolowski* The Baylor College of Dentistry and "The Sammons Cancer Center at Baylor University Medical Center, Dallas, Texas, USA Received June 199, and in final form 5 October 199 There are no data relating complex film tomography with effective dose that may be used to estimate the relative risk associatedwith dental implantdiagnostics. The purpose of this study was to calculate the effective dose and estimate risk from the use of the Scanora multimodal imaging system. With the use of a tissue equivalent humanphantom and thermoluminescent dosimetry, panoramic radiography was found to result in an effective dose of psv, while complex film tomography resulted in an effective dose of <1 psv to 0 psv depending on the anatomical locationof the imaging plane and the collimation option. An effective dose of this magnitude for panoramic radiography was estimated to represent a probability for stochastic effects on the orderof 1.9 x 1O--{;. Similarly, the effective dose associatedwith film tomography may be estimated to be equal to a probability for stochastic effects in the range of ~1 x 10--{; to. X 1O--{;. Keywords: Radiography, panoramic; radiography, dental; tomography, X-ray; radiation, dosage Dentomaxillofac. Radio!., 1994, Vo!., 1-1, August A number of parameters, such as the skin entry dose (surface exposure), surface exposure integral, absorbed dose, and the integral absorbed dose or energy imparted to the patient have been used in attempts to place the risk or detriment from low levels of partial body X-ray exposure, such as encountered in diagnostic radiography, in perspective. None of these methods consider either the radiosensitivity of the organs or tissues exposed or their relative contribution to the well-being of the body as a whole. To overcome these deficiencies, the International Commission on Radiological Protection in ICRP (19)1 defined the relationship between the probability of stochastic effects in certain defined organs and tissues, and a unit of radiation dose termed the effective dose equivalent (HE)' This unit was later modified to create the effective dose (E) in 1990 in ICRP &Y. Detriment is a term currently used in ICRP <Y to represent a combination of the probability of occurrence of an unfavourable health effect resulting from radiation exposure and a judgment as to the severity of the effect. Thus, radiation detriment may be expressed as the total harm that would eventually be experienced by an exposed population of individuals and their descendants as a result of radiation exposure. This includes the probability of fatal cancer, the weighted probability of non-fatal cancer, the weighted probability of hereditary effects, and the relative length of life lost. At present, the whole population probability 1994 Butterworth-Heinemann for the IADMFR 0SG-8X/94/001-0S coefficient for these stochastic effects resulting from exposure to low doses of radiation is. X 10- Sv- 1 This figure includes an estimate of 5.0 x 10- Sv- 1 for fatal cancers, 1. x 10- Sv- 1 for hereditary effects, and 1.0 x 10- Sv- 1 for non-fatal cancers. Detriment is measured by E which represents the doubly weighted sum of the absorbed doses to defined tissues and organs of the body calculated as follows. The equivalent dose (H T ) is the sum of the products of the radiation weighting factor (WR), which is unity for X-rays of all energies, and DT,R which is the absorbed dose averaged over tissue or organ T, due to radiation R, such that (Equation 1) E is the sum of the equivalent doses to each organ or tissue multiplied by that organ or tissue's weighting factor (WT), such that E =IHTwT (Equation ) Tissue weighting factors represent the relative contribution of that organ or tissue to the total detriment resulting from a uniform whole body irradiation. The effective dose calculated in this manner may be expected to result in the same total detriment as that from a uniform whole body exposure of the same magnitude. Thus, effective dose allows comparisons to be made between radiographical techniques which Dentomaxillofac. Radiol., 1994, Vol., August 1

2 expose only portions of the body and whole body exposures resulting from natural or background radiation. Whether factors have been assigned to both specific organs and tissues, and to a group of organs and tissues called the remainder. The remainder is composed of those which are likely to be selectively irradiated. Some of the remainder are also known to be susceptible to radiation-induced cancer. The salivary glands are not included in ICRP 0 with either a specific weighting factor or as part of the remainder. However, for this study, the salivary glands were assigned to the remainder and calculation of E modified to reflect their inclusion. The decision to treat the salivary glands in this manner was based not only on the fact that these glands were felt to be selectively irradiated during oral radiographical studies, but also on the reported prob-ability of fatal cancer induction in the salivary glands of the order of 5 x 1<r4sv',. This value is equal to the probability of fatal cancer induction on bone surfaces, which is a weighted tissue according to ICRP 0. This treatment of the salivary glands is consistent with the ICRP, which states that other tissues or organs either selectively irradiated or later identified as having a significant risk of induced cancer will be included, either with a specific weighting factor or in the list constituting the remainder. Because effective dose is a relatively recent calculation and because multiple, non-standardized models have been used, data in the literature demonstrate a wide range of values. A recent review by White took these variables into consideration in the process of developing an assessment of radiation risk from oral radiography". He calculated a mean E from published data for a full-mouth examination made with round collimation and D-speed film of 84f-LSV with a range of 4-14f-LSv and a mean E for panoramic radio-graphy of.f-lsv, less than 10% that of a full-mouth survey. At present, there is no data relating complex film tomography with effective dose. This study was therefore undertaken to establish the effective dose from this technique, so as to establish a level of relative risk associated with its use for dental implant diagnostics. Materials and methods Dosimetry was performed using thermoluminescent dosemeters (TLD). All TLDs (TLD-100 lithium fluoride extruded rods, 1mm x 1 mm x mm, Solon Technologies, Inc., Solon, OH, USA) were from a matched extrusion lot, and subjected to a selection process in which those dosemeters whose response varied by more than 10% from the group's mean were discarded. TLDs were exposed to 90kVp X-rays (Model 1000, HVL,. mm Al equivalent, Gendex, Milwaukee, WI, USA) and radiation intensities were measured with a Radcal 1015C radiation monitor (MDH Industries, Inc., Monrovia, CA, USA) using a cm air equivalent volume ionization chamber. Prior to use, all TLDs were annealed at 400 C for 1h followed by h at 100 C. Dosemeters were read 4h Table I Dosemeter location Organ/tissue Thyroid Salivary glands Parotid (right) Parotid (left) Submandibular (right) Submandibular (left) Bone marrow Third molar (right) Third molar (left) First molar (right) First molar (left) First premolar (right) First premolar (left) Mandibular symphysis Second cervical vertebra Sixth cervical vertebra Lateral calvarium (right) Lateral calvarium (left) Posterior calvarium Anterior calvarium Oesophagus Skin Philtrum Occipital Vertex Preauricular (right) Preauricular (left) Posterior neck Chin Brain Phantom level Level number of tissue-equivalent human phantom (Alderson Research Laboratories, Stamford, cr, USA) after exposure, to allow for the decay of short half-life glow peaks, using a recently calibrated Harshaw Model 000AIB thermoluminescence detector/automatic integrating picoammeter (Solon Technologies, Inc., Solon, OH, USA) operated according to the manufacturer's instructions. The identity of all the selected dosemeters was maintained throughout the study to permit the use of individual calibration factors for conversion of the readings obtained from the Model 000AIB in C x 10-9 to R. For each radiographical projection studied, 4 TLDs were placed in sites representing weighted tissues or organs in the upper portion of a tissue equivalent human phantom, which consisted of 15.5 cm horizontal sections (Alderson Research Laboratories, Stamford, CT, USA). These sites were selected to record the exposure to the skin, and the mean absorbed dose to bone marrow, thyroid, pituitary (representative of the brain), parotid and submandibular glands, and the oesophagus (Table I). Additionally, at least three TLDs were used to record background. The phantom including the TLDs was positioned in a Scanora'P X-ray unit (Orion Corporation/Soredex, Helsinki, Finland) and exposed to 40 imaging cycles using average adult techniques (patient size 5). For the panoramic study, these techniques were kvp and 40mAs, for the anterior tomographic study kvp and 84 mas, for the premolar tomographic study kvp and 1OmAs, and for the molar tomographic study kvp and 18mAs. The use of T-MAT G film and Lanex medium screens (Eastman Kodak Co., Rochester, NY, USA) was simulated for all studies, o 8 14 Dentomaxillofac. Radiol., 1994, Vol., August

3 except the anterior tomographic which used Lanex fine screens. The selection of screen speed was basedon the manufacturer's operational directions (Orion Corporation/Soredex, Helsinki, Finland). Octospiral tomograms (layer thickness 4mm) centred on the occlusal plane were made of the three areas using both the rectangular (0 x 10mm [.14 x 1(fmm/[) and the smaller round (5 mm [.55 X 10 mm ]) collimation options of the X-ray unit. Following exposure, readings of the dosemeters in C X 10-9 were divided by 40, converted to R using the above-mentioned calibration factors, and R converted to Gy using the factor 8.9 x 10-,5. The absorbed dose recorded at each of the sites.represents the mean of - dosemeters at each site, corrected for background. In this study absorbed dose was equated with equivalent dose. For all calculations the marrow content of the mandible, calvarium, and cervical spine was considered to be 1.5% of total body marrow (11.8% calvarium, 1.% mandible,.4% cervical vertebra)", the bone surface of the skull and cervical spine 15.5% of total bone surface (14.1% skull and 1.4% cervical vertebra)", and the skin of the head and neck 9.0% of the total skin surface of the body", Absorbed dose in cortical bone was calculated by multiplying the mean marrow dose by 4.4, which is the ratio of f factors (the factors for conversion of exposure to absorbed dose) for bone and soft tissue". Effective dose was calculated according to Equation. Weighted organs not included in this study were the gonads, colon, lung, stomach, bladder, breast, and liver, because of their small contribution to the effective dose equivalent calculated from panoramic radiography'", For the same reason, the brain (as the pituitary gland) was the only remainder organ included. All others were assumed to receive an equivalent dose of zero. With the addition of the salivary glands as part of the remainder, the total number of remainder organs became 11 for purposes of calculation. In those cases where the salivary glands received an equivalent dose in excess of the highest dose to any of the measured weighted organs, a weighting factor of 0.05 was assigned to the salivary glands and a weighting factor of 0.05 applied to the average dose to the 10 remainder The nominal probability coefficient for stochastic effects in a population of all age groups of. x 10- Sv- 1, was used to calculate the probability for the induction of fatal and non-fatal cancer, and severe hereditary effects. Results The E for the techniques studied is shown in Tables II-V. Panoramic radiography (Table II) was found to deliver an E of f.lsv. Because the salivary glands received an equivalent dose which exceeded thatof any of the measured weighted tissues or organs (5f.LSv), they were assigned a weighting factor of.5 x 10- or one half the remainder. The salivary glands were treated in a like manner in tomography of the premolar Table n Panoramic radiography! Equivalent Effective Weighting dose dose Site factor: (u Sv) (p.sv) Bone marrow Oesophagus Thyroid Skin <l11 Bone surface Salivary glands Other remainder <1 5 Total 1Equivalent doses and effective dose represent the mean of determinations. ICRP Publication 0. Radiation protection Recommendations of the International Commission on Radiological Protection. Oxford: Pergamon Press, <el= <0.1. 4The equivalent dose to the brain was OpSv. This value (psv) represents the mean of 10 remainder 5<1 = Table m Tomography anterior midline collimation] '0xl0mm. 5mm (diameter). Weighting factors. Reference. 4<1 = <el= <0.1. HT to the brain was 14pSv and to the salivary glands 49pSv. This value (psv) represents the mean of 11 remainder HT to the brain was 4pSv and to the salivary glands 8pSv. This value (lpsv) represents the mean of 11 remainder collimation] collimatiorr Site WR (u.s») (p.sv) (p.sv) (p.sv) Bone marrow <1 4 Oesophagus <1 Thyroid <1 Skin <l11 5 <1 Bone surface <l <1 1 <l11 Total 5 <1 Table IV Tomography premolar area Site WR (JioSv) (JioSv) (JioSv) (JioSv) Bone marrow <1 4 Oesophagus Thyroid Skin <l11 Bone surface <1 Salivary glands Other remainder 0.05 <1 <l11 Total 0 1 '0x l0mm. 5mm (diameter). Jweighting factors. Reference. 4«1 = <el= <0.1. HT to the brain was 55pSv. This value (psv) represents the mean of 10 remainder HT to the brain was 5pSv. This value (psv) represents the mean of 10 remainder and molar areas, where they again received an equivalent dose in excess of that received by any other organ or tissue measured (Tables IV and V). The E from a single anterior midline tomogram was 5f.LSv when rectangular collimation was used (Table III). This was reduced to <If.LSv with smaller round Dentomaxillofac. Radiol., 1994, Vol., August 15

4 Table V Tomography molar area Site WR (psv) (psv) (psv) (psv) Bone marrow <1 4 Oesophagus <1 Thyroid Skin <ali15 <alii Bone surface <1 Salivary glands Other remainder 0.05 ' <1 1 <alii Total 11 '0x10mm. 5 mm (diameter). 'weighting factors. Reference. 4<1 = <0.1-4l.9. 51= <0.1. HT to the brain was 9j4Sv. This value (j4sv) represents the mean of 10 remainder HT to the brain was 14j4Sv. This value (lj4sv) represents the mean of 10 remainder Table VI Probability of stochastic effects by technique! Technique Panoramic Film tomography by location and collimation/ Anterior Premolar Molar Probability (x10--) 'The whole population probability coefficient is. X 10- Sv-'. Reference. collimation, 0 x 10mm, round collimation, 5mm (diameler). 1 = <0.1. collimation. A similar pattern of dose reduction was found in the premolar and molar area tomograms (Tables IV and V). The probability for stochastic effects resulting from exposure by these techniques is shown in Table VI. It can be seen that a panoramic radiograph corresponds to a risk of 1.9 x 1O~ and the tomographic images range from ~1-. x 1O~, depending on the area imaged and the collimation employed. Discussion Originally proposed for radiation protection purposes, the E has been used almost since its introduction as a means by which comparative risk estimates in diagnostic radiography can be established. The method used for calculating E in this study was modified from that described in ICRP 0 to include the salivary glands as part of the remainder This method will result in a higher estimation of E, the magnitude of which can be seen in the data. The calculated E from panoramic radiography was p,sv. If the salivary glands were ignored, the recalculated E would be lop,sv, 8% lower and closer to the mean of.p,sv calculated by White 4. Likewise, the data show that the salivary glands contribute from 0-50% of the E calculated from the tomographic studies, depending on the area imaged and the collimation option. Equivalent doses recorded by the TLDs were assumed to represent the mean tissue or organ dose. These equivalent doses and calculated effective doses were dependent, in part, on the location of some organs in relation to the image plane, as defined by the location of the dosemeters, and on the inhomogeneous nature of dose distributions in oral radiographical exposures'", Thus, the equivalent dose to the salivary glands (combined parotid and submandibular) from an anterior tomogram using rectangular collimation was 49p,Sv, from a premolar area tomogram 98p,Sv, and from a molar area tomogram 5p,Sv. The areas imaged by these procedures were anatomically anterior, equal to, and slightly posterior to the major salivary glands respectively. This same relationship was found when the smaller round collimation option was used. This finding was not unexpected, as similar results have been reported in the literature!'. The relatively high equivalent dose found for the skin in anterior tomography using round collimation (p,sv) may have resulted from a fortuitous alignment of the centre of the image plane with those TLDs placed in the area of the philtrum. With this exception, the equivalent dose to the bone marrow and skin was relatively constant, regardless of the anatomical location of the image plane due to the distribution of bone marrow and skin within the body. The relationship between the location of the organs and the tomographic image plane, and the lack of homogeneity of the X-ray beam, also appeared to influence the magnitude of dose reduction realized by using the round collimation option. The use of the round collimator might be expected to result in an E of % of that of rectangular collimation, based on the cross-sectional area of the X-ray beam. While the anterior tomograms made with round collimation delivered an E of aboutonly 0% of that of rectangular collimation, those made of the premolar and molar areas were nearly twice as much higher. While the probability of stochastic effects resulting from panoramic radiography was almost four times higher than the value calculated from the literature of 0.5 x 10~,4, it must be remembered that this data includes the salivary glands which were responsible for 5% of the E. In considering the clinical application of this data to implant diagnostics with the Scanora'ssystem, it should be appreciated that, prior to tomography, a panoramic radiograph is needed to establish the coordinate of the area to be examined 1. Once this coordinate is determined, and the specific tomography program selected, three or four tomographs are made during each imaging cycle. While this method is clinically useful for the evaluation of multiple contiguous fixture sites, it should be clear that imaging by this system will result in a total E equal to the panoramic image plus three to four times that reported for a single tomogram. Multiple tomographic imaging per implant site is not unique, and is often used clinically with other 1 Dentomaxillofac. Radiol., 1994, Vol., August

5 tomographic systems':'. Additionally, if only a single tomogram is required, the Scanoraf system may be manually overridden. Inflation of the risk associated with maxillofacial radiography because of the inclusion of the major salivary glands in calculations must be added to that already ascribed to the effective dose when calculated for inhomogeneous dose distributions'p'!". In effect, however, this ensures that this method of reporting the risk is consistent with a conservative approach to radiation safety. Acknowledgements This work 'was supported in part by a grant from Soredex Medical Systems of Conroe, Texas USA. The authors would like to thank Dr. Barton M. Gratt, UCLA School of Dentistry, Los Angeles, California USA for use of the TLD equipment. References 1. ICRP Publication. Radiation protection. Recommendations of the International Commission on Radiological Protection. Oxford: Pergamon Press, 19.. ICRP Publication 0. Radiation protection Recommendations ofthe InternationalCommission on Radiological Protection. Oxford: Pergamon Press, UNSCEAR 19. Sources and effects of ionizing radiation. United Nations Scientific Committee on the Effects of Atomic Radiation. 19Report to the General Assembly with Annexes. New York: United Nations, White SC. 199 assessment of radiation risk from dental radiography. Dentomaxillofac RadioI199; 1: ICRU Report 1. Radiation dosimetry: X-rays generated at potentials of 5 to 150kV. Washington D.C.: International Commission on Radiation Units and Measurements, Ellis RE. The distriution of active bone marrow in the adult. Phys Med Bioi 191; 5: ICRP Publication. Report on the task group on referenceman. International Commission on Radiological Protection. Oxford: Pergamon Press, Boswick JA. The art and science of burn care. Rockville, Maryland: Aspen Publishers, Inc., Johns HE, Cunningham JR. The physics of radiology. 4th ed. Springfield, Illinois: Charles C. Thomas, Gibbs SJ. Influence of organs in the ICRP's remainder on effective dose equivalent computed for diagnostic radiation exposures. Health Physics 1989; 5: Bristow RG, Wood RE, Dark GM. Thyroid dose distribution in dental radiography. Oral Surg Oral Med Oral Patho11989; 8: Tammisalo E, Hallikainen D, Kanerva H, Tamrnisalo T. Comprehensive oral X-ray diagnosis: Scanora multimodal radiography. A preliminary description. Dentomaxillofac Radial 199; 1: Kassebaum DK, Stoller NE, McDavid WO, Goshorn B, Ahrens CR. Absorbed dose determination for tomographic implant site assessment techniques. Oral Surg Oral Med Oral PathoI199; : Velders XL, van Aken J, van der Stelt PF. Risk assessment from bitewing radiography. Dentomaxillofac RadioI1991; 0: Address: Dr Neil L. Frederiksen, The Baylor College of Dentistry, 0 Gaston Avenue, Dallas, Texas 54, USA. Dentomaxillofac. Radiol., 1994, Vol., August 1