Final Report Summary - ORAMED (Optimization of radiation protection of medical staff)
Executive summary:
The medical staff that carries out interventional radiology and cardiology (IR/IC) procedures is likely to receive significant radiation doses to their hands, or parts of their body not covered with protective equipment, as they are close to the x-ray field. The dose range for the same kind of procedure varies a lot, since there are many factors affecting extremity doses. In addition, there is evidence that eye lens doses can be high in IR/IC and cases of lens opacities and cataract have been reported in recent years. Besides, a lack of appropriate equipment is also identified in the field of active personal dosimeters (APDs) for typical fields in IR/IC. Very few devices can detect low energy fields and none of them are really designed for working in pulsed radiation fields. In the field of nuclear medicine (NM) the extremity doses to the medical staff are also known to be very high. One can highlight the difficulties in estimating the dose distribution across the hands and there is a need for better knowledge of doses received during the main tasks of a NM department, especially using unsealed sources.
The ORAMED project, (see http://www.oramed-fp7.eu online) was set up to optimise the working procedures in these medical fields with respect to radiation protection. The project was structured in five work packages (WPs):
1. extremity and eye lens dosimetry in IR and IC
2. development of practical eye lens dosimetry
3. optimisation of the use of APDs in IR and IC
4. extremity dosimetry in NM
5. training and dissemination.
A coordinated measurement program in European hospitals was organised both in IR/IC and NM departments. Moreover, simulations of the most representative workplaces/procedures were performed to determine the main parameters that influence the extremity and eye lens doses. Next to these, some dedicated studies on improving the eye lens dosimetry and active personal dosimetry were conducted.
Based on the measurement and simulation results, a series of practical guidelines and training packages were developed. The influence of the different radiation protection measures, such as lead (Pb) shields, has been quantified and clear monitoring requirements have been formulated for a series of medical procedures.
A formalism for the use of the operational quantities for eye lens dose measurements has been worked out (calibration phantom, conversion coefficients, type test procedures). A dedicated eye lens dosimeter has been developed during the project and is now commercially available to be used in the hospitals.
Different existing APDs have been tested for the first time under conditions that are representative of the conditions met in clinical practice (e.g. pulsed fields, low energies) and a series of guidelines for the use of these APDs in IR/IC have been made. Moreover, an improved APD device designed specifically for IR/IC fields has been developed.
The extensive measurement and simulation campaign for extremity doses in NM lead to a systematic evaluation of the different radiation protection measures. These are condensed in a series of practical guidelines to be used. The dose distribution across the hands of the technologists and the physicians was characterised and recommendations for routine monitoring have been formulated.
The outcome of the ORAMED project will lead to an optimisation of the radiation protection standards for medical staff. The practical guidelines that have been developed will be used by medical staff in hospitals for many years to come. In particular, the developments on the eye-lens dosimetry and the APDs will improve the practical measurement capabilities in the field.
Project context and objectives:
The staff that carries out interventional procedures is likely to receive significant radiation doses to their hands, or parts of their body not covered with a protective apron, as they are close to x-ray tubes. The doses for this kind of procedures vary a lot, since there are many factors affecting extremity doses such as protective devices, x-ray geometry and spectra, the scattered radiation from the patient, etc.
There are cases mentioned in the literature where the extremity doses can approach the dose limits. In these cases either the high workload or the lack of a proper radiation protection policy are responsible for the high doses observed.
Routine monitoring of extremities is difficult, since 'the most exposed area' according to the International Commission on Radiological Protection (ICRP) recommendations cannot easily be found. In most cases only finger or hand doses are reported; doses to the eye lens, legs or thyroid have not been evaluated. In some studies doses to the legs can be even higher than doses to the hands. Even when ring/wrist dosimetry is used for extremity monitoring the position of the dosimeter is not clear. There is evidence that eye lens doses are high in interventional radiology and cases of lens opacities and cataract have been reported in recent years. However, eye lens doses are never measured in routine applications and in addition very few data can be found in the literature. There is no suitable dosimeter available and the standards for the operational quantity measurements are not complete. This situation is partly due to the lack of conversion coefficients and a suitable calibration procedure.
A lack of appropriate equipment is also identified in the field of APDs for typical fields in IR/IC. Very few devices can detect low energy photon fields and none of them are really designed for working in pulsed radiation fields.
The literature concerning radiation exposure and protection of NM staff is much more limited than in the case of interventional radiology and mostly refers to conventional diagnostic NM.
As a consequence of the definition that the dose limit for the skin has to be applied to 'the dose averaged over any area of 1 cm2 regardless of the area exposed' it is advisable to measure the local skin dose at the location with presumably the highest exposure. This requirement is the central dilemma of extremity dosimetry and causes severe practical difficulties. In daily practice when preparing and administering radio-pharmaceuticals in NM it is not easy to comply with that requirement since it is often not known which part of the hand receives the highest dose. Moreover, the dose distribution over the hand may vary during a single process as well as when various workers perform the same procedure.
Unsealed radiation sources are being increasingly used in NM for radiation therapy, in particular, nuclides that emit beta or mixed beta/gamma radiation. Considering the preferential use of beta emitters, the dosimeters must be appropriate for beta radiation, taking into account both the energy spectra of the nuclides and the spectral dose response of the dosimeter. This means that in NM therapy, staff may be exposed to high doses, even exceeding the annual limit of the dose to the skin of 500 mSv. Thus, adequate safety measures including an accurate individual monitoring of the personnel are a strict requirement.
It should be pointed out that most general official personal monitoring databases, such as the United Nations Scientific Committee on the Effects of Atomic radiation (Unscear) or the Information System on Occupational Exposure (ISOE), do not include extremity doses and that many countries with NM services do not have authorised ring dosimetry services. Furthermore, the data available in official national registers are much lower than the estimated doses in research studies, which probably means that monitoring is not being done correctly. Through the dissemination of the ORAMED results and the training, we will aim at the improvement of the radiation protection in these fields.
The ORAMED project proposes to develop methodologies for better assessing and reducing exposures to medical staff. This general objective is being achieved through the development of five main topics, structured in five WPs.
1. The objective of WP1 was to obtain a set of standardised data on doses for staff in IR/IC and to optimise staff protection.
2. The objective of WP2 was to establish a sound theoretical and experimental basis to assess eye lens doses.
3. The objective of WP3 was to optimise the use of APDs in interventional radiology.
4. The objective of WP4 was to detect the most exposed part of the skin by measuring the extremity doses and dose distributions across the hands of the medical staff working in NM departments.
5. The last objective of the project (WP5) was to design and develop an accurate teaching and knowledge dissemination program and to make sure that the conclusions and recommendations of the project are transmitted to the stakeholders, mainly medical staff, radiation protection officers, dosimetry services and authorities in the field.
Project results:
WP1: Extremity dosimetry and eye lens dosimetry in IR/IC
Measurement campaign
List of procedures
The list of procedures includes three cardiac and five general interventional diagnostic and therapeutic examinations. More specifically, the list is composed of cardiac angiographies (CA) and angioplasties (PTCA), radiofrequency ablations (RFA), pacemaker and cardiac defibrillator implantations (PM/CD), angiographies (DSA) and angioplasties (PTA) of the lower limbs (LL), the carotids and the brain (C/B) and the reins (R), embolisations and endoscopic retrograde cholangiopancreatographies (ERCP). The choice of the procedures was based on their potential impact on the annual exposure of the staff. Thus two main selection criteria were defined: high annual frequency and possible high kerma area product (KAP) values. However, some procedures of low frequency were also considered in the study, in order to include the different parts of the patient body that is irradiated, or because they are often performed in rooms with limited shielding equipment.
Measurement protocol
A measurement protocol was established, in order to obtain an accurate description of the staff irradiation scenario and to homogenise the collected data. Thus, different parameters related to the angiographic system, the type and complexity of the procedure, the position of the physician, the protective equipment, the experience of the physician, some field parameters and finally the fluoroscopy time, the number of images acquired and the KAP values were recorded.
For the measurements it was decided to use high sensitivity thermoluminescent dosimeters (TLDs). The TLDs were sealed in small plastic bags and taped on the parts of the body to be monitored.
Results
Among the types of cardiac procedures that were included in the measurement campaign, the doses to the operators are higher during the pacemakers and implantation of cardiac defibrillators (PM/ICD), even though the respective KAP values are relatively low since only fluoroscopy is used. During these procedures the operators work very close to the irradiation field and most of the time without any protective shield.
Among the IR procedures special attention should be given to embolisations, particularly to the doses to the eye lenses. Operators are also significantly exposed during therapeutic procedures such as angioplasties of the lower limbs and the renal arteries. During cerebral and carotid procedures the doses are relatively low since femoral access is usually used and the operator stands at a larger distance from the irradiated part of the patient's body compared to other procedures performed in the thoracic or abdominal region.
Finally, for ERCP procedures the doses are generally low. Special care should be taken regarding the use of a ceiling suspended shield, especially for the protection of the eyes, when overcouch irradiation is used.
The position of the maximum dose was found to be at the part of the operator which is close to the x-ray tube. More generally, in all the procedures studied the left wrist was found to be the position where the maximum dose was recorded most often, followed by the left finger. The maximum was found generally at the left wrist for femoral access and at the left finger for radial or direct access, the operator being then closer to the beam field. Therefore, both wrist and ring dosimeters are suitable for the routine monitoring. However, taking into account the respective annual limits for all positions the eye lens exposure is more important.
According to the information on the annual workload that was gathered from 84 physicians, the annual limit for the skin on the fingers could be exceeded only for a few workers but the annual doses were estimated based only on the procedures for which the operator was monitored during the measurement campaign.
In the case of the CA/PTCA procedures, 5 to 15 % of the workers seem possible to receive an annual dose above the three tenths of the annual limit. Furthermore, there are around 5 % of the workers that could even exceed the annual skin effective dose of 500 mSv for PM/ICD procedures. Thus, the monitoring of the hands is definitely a requirement for interventional procedures especially in interventional cardiology. The annual doses were considered low for the ERCP procedures, so the monitoring of the hands is not necessary in this case.
Although the annual doses to the legs can be high for certain kind of procedures, it was shown within the analysis performed in WP1 that this was mainly happening for tube, below couch configurations without any shield attached to the operating table. Therefore, the monitoring of the doses to the legs is not needed when a table shield is properly used.
Finally, the eye lens doses can be high so monitoring is also recommended for all investigated procedures, except ERCP. Especially, if the dose limit for the eye is decreased in the future, monitoring and protection measures will be even more important.
Simulations
Description of the input file
The numerical simulations have been performed using the Monte Carlo n-particle extended (MCNP)-X code. The medical internal radiation dose (MIRD) type anthropomorphic models have been used for the simulation of the patient and the operator. The 'patient' phantom is at supine position and the 'operator' one is standing close to it, in a configuration that is typical in an IR and IC procedure. The original model was modified in order to represent more realistically the irradiation scenario: eyes have been added, the arms have been redefined and forearms and hands are bent in a more realistic position. A thyroid collar of 0.5 mm Pb and an apron of 0.5 mm Pb in front of the body have also been added. Finally, a cell filled with air representing the KAP chamber and the image intensifier have been added to the input file. F6 tallies were used for the calculation of the doses to the eye lenses, hands, wrists and legs. The tallies were positioned at 0.07 mm and at 3 mm depth for the calculation of Hp(0.07) and Hp(3) respectively.
Geometry characteristics
The x-ray tube was simplified to a point source emitting a photon beam in a cone with an aperture corresponding to that desired for the simulation of the x-ray field diameter. The energy spectra for the selected kVp and filtrations were determined using the x-ray data of the Institute of Physics and Engineering in Medicine (IPEM), report 78. Moreover, the Pb collimator was not simulated explicitly, but defined as a volume killing all the photons entering inside it.
The first parameters examined within the simulation campaign are the tube voltage and the filtration. More specifically, the tube voltage was changed from 60 to 110 kVp and the filtrations from 3 to 6 mm aluminium (Al) and from 0 to 0.9 mm copper (Cu). Moreover, simulations also were performed to examine several cases of beam projections, field size, different positions of the operator, protective glasses for the operator, protective equipment for the room, and four different parts of the body being irradiated, namely head and neck, thorax, abdomen/pelvis and lower limbs.
Results
The beam projections for which the operator is the most exposed are the left anterior oblique (LAO), where the operator stands at the side of the X-ray tube and the cranial projections, where the image intensifier is towards the patient's head. When comparing the overcouch (AP) with undercouch (PA) irradiations, the doses are found significantly lower for the undercouch one, up to 12 times for the eyes, eight times for the hands and four times to the wrists, for the selected geometries. In case of PA the dose to the legs can be relevant if no table shield is used. For the lateral projections, the right anterior oblique (RAO) should be preferred, as the doses were found lower compared to the LAO, up to three times for the eyes, three times for the hands and 22 times for the wrists, for the selected geometries.
The table shield is very effective for the protection of the legs and reduces the dose from 83 % to 99 % for the geometries and setups that were tested. The Pb glasses, for the specific geometry that was examined, are very effective for the protection of the eye lens which is closer to the x-ray tube. In practice, Pb glasses that cover well the eyes and also have lateral protection are advised. Lens thickness of more than 0.5 mm Pb doesn't improve the protection of the eye lenses significantly.
A ceiling suspended shield is also very effective for the protection of the eye lenses; its use is essential especially when Pb glasses are not available or cannot be worn for practical reasons and when overcouch irradiation is used. This type of shield also provides protection to the hands and wrists. The simulations showed how important it is to correctly place the ceiling shield. Even a small gap between the patient and the shield reduces the effectiveness of the shield to the hands. For this reason a shield with Pb stripes attached at the bottom so that it touches the patient is advised. For the lateral projections the eyes are better protected when the shield is positioned closer to the side of the operator and not just above the patient. In practice, a second shield is advised especially when biplane systems are used, so that the operator is protected from both x-ray tubes. For all monitored positions the doses are higher when larger field sizes are used.
Concerning the beam quality, more energetic beams lead to lower doses to the operator up to 60 %. However, one should keep in mind the deterioration of image quality and contrast in this case.
The measurement and simulation campaign performed within the ORAMED project revealed a large variability of practices followed in different hospitals. As a consequence, the measured doses, even for specific procedures, vary significantly from one case to another. On the other hand, the simulation data showed the way that each parameter separately can influence the extremity and eye lens doses and not in combinations as it is the case from the measurement results. The combined data led to the following recommendations. It is important to note that some of the proposed guidelines cannot easily be adopted since there are restrictions from the medical point of view. However, some of them are easily adjustable and can improve the protection of the medical staff significantly:
1. the equipment used for IR/IC should fulfil specific requirements and standardisation in their design, manufacture, acceptance and maintenance
2. personal protective equipment should be used for all the personnel in the room
3. the ceiling suspended shield should be placed just above the patient, especially in the cases that the tube is above the operating table; the operator should stand well behind it
4. when ceiling suspended shield is not available or cannot be used for practical reasons protective Pb glasses should be used
5. the table shield should be always properly adjusted to protect both legs
6. the tube should be placed below the operating table when C-arm systems are used. However, the increase at the leg doses for this setup has to be compensated by the use of a properly positioned table shield.
7. if biplane systems are used, the proper use of lateral shield is very important for the protection of the eyes
8. mobile floor shield should be used for the assisting personnel that need to be in the irradiation room
9. the femoral access of the catheter should be preferred, if it is possible from the medical point of view, compared to the radial one
10. the use of an automatic image injector which allows the operator to leave the room during the image acquisitions is a practice which can reduce the doses significantly, especially to the hands
11. the operators should avoid direct exposure of hands to primary radiation
12. monitoring of the eyes and fingers should be performed on routine basis. The dosimeters should be worn on the side of the operator which is closest to the x-ray tube. The finger (or wrist) dosimeter should be placed on the dorsal or palmar side of the hand when the x-ray tube is placed above or below the operating table, respectively.
WP2: Development of practical eye lens dosimetry
The basic step of WP2 was to revise the approach for the definition and calculation of conversion coefficients for Hp(3). Relying on morphological studies on the head, it was decided to propose a cylindrical (20 cm diameter and 20 cm height) simplified four-element tissue equivalent phantom on which to calculate the operational quantity.
The discussion was especially motivated by the fact that the possible adoption of the trunk phantom also for calculating Hp(3) seemed to be rather inappropriate due to its strong deviation from the reality in terms of mass and shape. In particular the 30 x 30 cm2 plane face is too large and of different shape than the head. Moreover, the edges play a very important role when looking at the angular dependence of the operational quantity. Such effects also influence the calibration procedure, if such phantom is adopted in the procedure, as its backscatter characteristics are not corresponding to what happens in practice to a target organ like the eye embedded in the head.
A new set of photon conversion coefficients were obtained. The calculations demonstrated the better suitability of the adoption of the cylinder to reproduce the organ equivalent dose (eye lens) in various irradiation conditions.
It has to be pointed out that, as expected the operational quantity as determined on the proposed cylinder are in satisfactory agreement with the limiting quantity. Moreover it should be noted that for nearly grazing incidence that could occur in IR/IC procedures, in which the patient acts as a scattering object, the operational quantity as determined in the trunk slab is not conservative estimate of HT(eye lens). The present general considerations justify the adoption of the new phantom to develop a self-consistent eye lens dosimetry procedure. It has to be pointed out that the new calculated operational quantity Hp(3) based on the cylinder is also better fitting with the HT(eye lens) values actually reported in ICRP-74 and ICRU-57.
RADCARD dosimeter prototype
On the basis of the studies on Hp(3) and the new set of conversion coefficients produced, a prototype was developed by RADCARD with the support of ENEA and CEA. The dosimeter had to measure correctly Hp(3) for eye-lens while being comfortable for users and for dosimetric services, waterproof and inexpensive. The characteristics to be optimised were the energy response and the angular response. The study was performed mainly varying the TLD detector type and dimension, the capsule material and dimension. The experimental irradiations were carried out using beams from the RQR and narrow series. The first series, with a moderate filtrationi better reproduce the operative condition met in IR/IC, whilst the second, with high filtration is closer to mono-energetic photons.
The prototype was finally tested at the CEA-LNE Laboratories in Saclay (France) and the irradiation results were compared with Monte Carlo simulations. The calculations were performed with the narrow spectrum series. The results are in a very satisfactory agreement.
Establishing a Protocol for type test and calibration of personal dosimeters in terms of Hp(3)
This part of the work was aimed at making a proposal for the type test and calibration of eye lens passive dosimeters especially designed to be used in IR/IC. Starting from the only one existing standard dealing with eye lens dosimetry using TLDs parameters such as, detection threshold, energy and angle responses have been reviewed and have been harmonised, as much as possible, with the International Electrotechnical Commission (IEC) 62387 requirements, taking into account the particular use at IC/IR workplaces. Conversion coefficients from air kerma to dose equivalent at 3 mm depth for RQR and International Organisation for Standardisation (ISO) radiation qualities, employed for type test and calibration purposes, have been calculated in the new phantom introduced by ORAMED.
Type tests are intended to demonstrate that the dosimeters are suitable for measurements in workplace conditions. For dosimetry systems based on passive personal dosimeters, two international standards covering photon and beta radiations exist for type testing: IEC 62387-1 and ISO 12794. The comprehensive document treats the following aspects:
1. wearing conditions at the IR/IC workplaces and consequences on type test criteria
2. detection threshold
3. performance requirements for energy response
4. performance requirements for angle response
5. other requirements
6. Calibration.
Conclusions
WP2 of the ORAMED project allowed establishing a new and self-consistent approach for an improved dosimetry procedure on the eye lens. Starting from the proposal of a well suited simple phantom for the theoretical quantity, the corresponding designed calibration phantom was fabricated at a very low cost. Thereafter the phantom was adopted during the ORAMED project for calibrations of the prototype. It was developed to respond in the best way in terms of Hp(3), according to the protocol that was conceived to offer clear guidelines specifically addressed to the eye lens personal dosimetry. The obtained results can be of help to the dosimetry community working in the medical field and also in other disciplines.
WP3: Optimisation of the use of APDs in IR/IC
Tests of eight commercial APDs in laboratory conditions with continuous and pulsed x-ray beams and in hospitals
The tests with continuous x-ray fields were made in two calibration laboratories. These tests were performed to determine the response of selected APDs in terms of personal dose equivalent, energy, personal dose equivalent rate and angle. The reference fields were used as defined in the ISO 4037-1 standard. Three measurements per APD were performed. Two dosimeters of each type were tested. Dosimeters were placed on an ISO slab phantom. The results were analysed considering the requirements of the IEC 61526 standard.
Results
The personal dose equivalent response of the tested APDs is linear in the dose range of interest in radiation protection, i.e. up to 500 mSv. The energy response of the tested APDs is within the interval as required in the IEC 61526 standard, from S-Co energy down to N-30 for all APDs except EDD30 and DoseAware. EDD30 energy response is within the IEC requirements between N-80 and N-20 qualities; dose Aware energy response is within requirements between N-120 and N-40. Most APDs can stand high dose rates up to 10 Sv.h-1 except PM1621A, for which the response diverges rapidly from 1 Sv.h-1 as well as EDD30 and DoseAware which saturate for personal dose equivalent rates above 2 and 4 Sv.h-1 respectively. It is interesting to notice that most APDs can stand personal dose equivalent rates higher than those indicated in their technical note. The angular response is within the interval for energies down to N-30 for all APDs, apart from AT 3509C for which the angular response is inside the before mentioned interval at 60 ° only for N-80.
Tests of APDs with pulsed x-ray beams
The tests in pulsed mode were performed at the French standard laboratory for ionising radiation. The influence of several parameters in different conditions on the response of the APD in pulsed mode was studied.
For most APDs, the response decreases when the personal dose equivalent rate increases. For personal dose equivalent rates lower than 2 Sv.h-1 the responses are, in general, close to one and fall down more or less rapidly for higher dose rates, except DIS-100 that gives a correct response up to 55 Sv.h-1. It was noticed that PM1621A, equipped with a Geiger-Muller tube, does not provide any signal at all in pulsed mode. DMC 2000XB, EPD Mk2.3 EDMIII, EDD30, AT3509C and DoseAware contain all silicon detectors. The difference of their response to the pulsed mode is probably due to the time response of their electronic systems. DIS-100, which has a 'hybrid' technology between silicon and ionisation chamber, presents correct results.
The variation of the APD response with the pulse frequency between 1 and 20 s-1 is roughly equal to 10 % for EDMIII, EDD30 and DoseAware and to 30 % for the other devices (except PM1621A).
When the pulse width is larger than 1 s, the responses in pulsed and continuous radiation field are quite similar. No significant effect of pulse width was observed on the response of all APDs. All results from the pulsed field tests show that the longer the pulses and the higher the frequency, the better the behaviour of the devices tends to be.
The first series of tests in real conditions were performed by positioning APDs on an ISO slab phantom representing the operator. The scattered irradiation was produced by an anthropomorphic Rando-Alderson phantom representing the patient. The tests were performed on an x-ray system (PHILIPS BZR79 Optimus). The APDs tested were: DMC 2000XB (MGPi), EPD Mk2.3 (Siemens), EDM III (Dosilab), PM1621A (Polimaster), DIS-100 (Rados) and EDD30 (Unfors).
As reference the routine passive TLD from the Belgian Nuclear Research Centre was used (uncertainty 20 %). Both the APD as the TLD were positioned together on the ISO slab phantom. The dose uniformity on the surface of the phantom was within 20 %. The thorax of the RA-phantom was irradiated and different realistic set-ups were considered. The main objective of these tests was the study of the behaviour of the APDs in realistic conditions with the possibility to select specific field parameters.
In a range of dose equivalent rates tested from 10 mSv.h-1 to 1.8 Sv.h-1 the APD response is within 50 % for the range of dose equivalent rates tested, except for the EDM III for which the dose is general higher than the TLD dose and except PM1621A that does not give any signal, which is consistent with the laboratory tests in pulsed fields. No important influence of the tube voltage and of the pulse width was observed on the APD response compared to the TLD.
Tests on operators
For these series of tests operators wore, side by side, one APD and one passive dosimeter above the Pb apron. The dosimeters were worn during several interventions to integrate doses of at least 300 µSv for several types of IR/IC procedures. The dose equivalent was provided by the passive dosimeter according to the routine measurement protocol of the respective partner that performed the measurements. For practical reasons only 5 APD types were tested: DMC 2000XB (MGPi), EPD Mk2.3 (Siemens), EDM III (Dosilab), DIS-100 (Rados) and DoseAware (Philips). In total 102 measurements were performed in seven different European hospitals. The main objective was to compare the measurements performed by the APD and passive dosimeter worn in routine practice where all kinds of procedures and parameter settings are used and without an accurate knowledge of these parameters.
With respect to passive dosimeters, in general all five tested APDs under-responded. We can observe a large spread in the results, which might be explained by non-uniform irradiations or the shielding of one dosimeter by the other.
Some recommendations were prepared within the group to help in selection and use of APDs at IR/IC workplaces. These recommendations are compiled on a three page leaflet available on the ORAMED website at http://www.oramed-fp7.eu/ They concern the selection of an APD in IR/IC and the use of an APD in IR/IC.
In addition, a prototype of an improved APD for IR was developed by MGPInstruments. The objective was to propose technical solutions to improve the response of an APD for an application in IR/IC based on the results of the tests in laboratory conditions. The different technical solutions developed by MIRION Technologies during this period were tested in laboratories. A prototype called 'OMEGA' using a new generation of ASIC 'CID 3G' was developed in order to replace the existing 'CID200' 1.2 µM CMOS process ASIC integrated in the DMC2000XB.
The first improvement on the OMEGA prototype consisted in having a good energy response. The second task was to reduce the angle influence on the dosimeter response. The position of the two detectors and the shape of the detection module have been re-designed in order to have a good isotropy for angles of irradiation greater than +/- 60°. In addition, a study of the dead time compensation was conducted in order to reach dose rates as high as 10 Sv.h-1.
WP4: Extremity dosimetry in NM
The objectives set out in WP4 were to evaluate extremity doses and dose distributions across the hands, to study the influence of protective devices and to propose 'levels of indicative reference doses' for each standard NM procedure, with the final goal of reducing, when possible, hand dose levels. One of the main tasks was to perform an extensive measurement campaign of hand doses in European hospitals using a unified measurement protocol.
The preparation and administration of both diagnostic and therapy procedures were studied. In diagnostics, radiopharmaceuticals labelled with 99 mTc and 18F were included in the investigations because of their wide use. For therapeutic procedures the studies were focussed on 90Y labelled radiopharmaceuticals.
For each measurement, preparation and administration to the patient of the radiopharmaceutical were separated. Eleven TLDs, calibrated to measure the personal dose equivalent Hp(0.07) were taped on gloves at 11 positions of both left and right operator's hands. For practical reasons in therapy a given pair of gloves was used for a single preparation. For diagnostics the gloves were used several times by the same operator in order to integrate significant doses in the TLDs. The identification of the hospital and operator were recorded, together with the manipulated activities, shield(s) used and any difficulty that could occur during the operations. For most of the operators the measurement was repeated five times. In diagnostics, those workers with less than four set of measurements were rejected. The TLDs used by the partners were of different types. An intercomparison exercise was done for reference to ensure consistency of the results between different partners and to select specific techniques for specific applications.
The statistical analysis was performed with 641 measurements, collected in around 20 NM departments per procedure from six European countries. For each radionuclide and procedure the analysis was performed independently. The normalised Hp(0.07) (µSv/GBq) measured at each of the 22 monitored positions were averaged over the series of measurements for each worker. These mean values were used for the analysis. The maximum normalised local skin dose was calculated as the highest of those 22 mean values.
Many parameters and steps affect the local skin dose at hands, especially for preparation. In addition there was a lack of information on potential parameters of influence, such as the operation time. Furthermore, the tools to reduce finger doses were sometimes used in some steps and not in others during a single measurement, or were either used differently from one measurement to another. The number of times that the activity was manipulated was not taken into account. In general, the fact that the measurements were not systematically watched or recorded on video could cause some important details to be missing. The problem is thus complex and therefore the analysis had to be kept simple. In spite of this, the study provides a good overview of the level of finger exposure from a wide range of working habits and working procedures.
Workers were classified according to their maximum dose, in increasing order. Some workers were considered as outliers. Those workers have practices that significantly differed from those of the majority, leading to very high or very low exposures. It has been observed that most of the workers receiving very high doses did not use shielding for vials and/or syringes. On the other hand, those with very low exposures used semi-automatic equipment, which is not a common practice yet.
Very large ranges of maximum doses were found for the same procedure. The preparation of radiopharmaceuticals involves higher finger doses per activity than for the administration, because the procedures are longer and there are more steps requiring manipulations of the vials/syringes with higher activities, some of them without shield. The dose distribution across the hands was also studied. For all procedures it was observed that the non-dominant hand usually receives higher doses than the dominant hand. For all diagnostics and therapy procedures, the index tip of the non-dominant hand is the position where the maximum dose is most frequently received, followed by the thumb of the same hand for almost all procedures. Less frequently, the same positions of the dominant hand were also found to be common positions with maximum dose. In the majority of the therapy cases, as for diagnostics, the tip of the index finger or the thumb on the non-dominant hand was found to receive the maximum dose, especially in those cases where the exposure was high.
In order to identify the parameters of influence on the skin dose, workers were classified into categories for those parameters for which information was available and whenever the amount of data in each category was large enough. For all these cases the Mann Whitney-U test was applied to analyse the differences between the skin doses received by workers within each of the categories. The results of the test did not show statistically significant differences between the doses received by experienced and beginner workers. On the other hand, the shield was found to be a very important parameter of influence both for the vial and for the syringe, causing the differences between the doses received when using shield and when not using it to be statistically significant. When feedback was given to workers after a measurement series, informing them of their exposures and with a discussion of bad practices in particular cases, a decreased dose was observed in the subsequent measurements. That is of particular importance for radiation protection optimisation because it demonstrates that the correction of the bad habits, that usually requires a small effort, could imply a very large spare in extremity doses to the operator, e.g. by using shielding and tools to avoid any direct contact of the fingers with the source.
Wrist or ring dosimeters are typically used for routine monitoring. Although there is not a harmonised criterion for the position of the ring dosimeter, in practice it is usually placed at the base of the index, middle or ring fingers of the dominant hand since these positions do not hamper work causing an underestimation of the maximum dose. The underestimation was assessed, in a first step, by calculating the correlations between the dose at all measuring positions and the maximum dose. Although the routine monitoring positions do not correspond to the position of the maximum skin dose, they can be used to estimate this quantity.
In a second step the ratios between the maximum dose and the dose at relevant monitoring positions and at the index tip were calculated. The calculation was made for each single measurement and then averaged over the set of measurements of each worker and for all diagnostics procedures separately, but the differences among procedures were not found to be relevant. Due to this reason and considering that NM workers are usually involved in more than one diagnostic procedure, the ratios were also calculated by including all data from all diagnostics procedures.
The mean ratios are significantly higher for the wrist positions. The lowest mean ratios were found for the index tip position. The ratios are also lower for the base of the index finger than for the base of the ring finger and lower for the non-dominant hand than for the dominant one. Thus, according to these results, the use of wrist dosimeters should be avoided because of a very high underestimation and a lower correlation to the maximum dose.
The annual dose of the monitored workers involved in diagnostic procedures for the ORAMED project has been estimated. For this estimation, only those procedures from which measured values, from the ORAMED measurement campaign, where available for a specific worker have been considered. Their workload and the activity manipulated per year for each radionuclide, was considered. Tthe estimated annual dose is above 150 mSv, i.e. three tenths of the annual limit, for 51 % of the workers and for 20 % of the workers the annual dose limit was exceeded.
It has to be noticed that the real situation is more complex since usually a given worker will not perform only one but several different procedures. Therefore, the annual maximum dose is likely to be an underestimation whenever the worker is actually involved in more procedures than those actually measured. The results of the estimation highlight the need to monitor the worker and to optimise the radiation protection standard in NM.
Monte Carlo simulations were employed to determine at what extent the range of doses evaluated during measurements could be considered 'intrinsically related' to those procedures and in which way it was possible to estimate the effectiveness of different adopted methodology and shielding, with such large variability of data. More than 200 simulations were performed.
Voxel models representing the hands of the operator during some selected steps of administration and preparation procedures were obtained through the elaboration of computed tomography (CT) scans of real phantoms. The radioactive source was simulated as a cylinder of proper volume representing the manipulated syringe or the vials. A sensitivity analysis was performed.
Simulations show that, depending on the radionuclides and on the voxel hand model, factors ranging from of 0.5 to 3 can be reached only by considering small displacements of the source with respect to the sensor positions. These fluctuations are of the order of those encountered in the analysis of the measurements results. Moreover, they showed the advantages concerning dose reduction of a correct use of the shielding during preparation of radiopharmaceuticals and of the convenience of employing forceps as an additional protective factor also in case of shielded sources. Concerning the shielding, large reduction factors can be obtained when using the appropriate shielding. With Monte Carlo simulations and dose mapping it was also possible to investigate at what extent the positions initially considered for the TLDs in the measurement program were suitable to estimate the maximum of the dose.
The following recommendations were derived from the observations and results of the WP4 of the ORAMED project:
1. extremity monitoring is a necessity in NM
2. the base of the index finger of the non-dominant hand with the sensitive part of the dosimeter placed towards the inside of the hand is the recommended position for routine extremity monitoring in NM
3. an estimate of the maximum dose to the hands can be obtained by multiplying the reading of the dosimeter worn at the base of the index finger of the non-dominant hand by a factor of six
4. shielding of vials and syringes is essential
5. the minimum acceptable thickness of shielding required for a syringe is 2 mm of tungsten for 99mTc and 5 mm of tungsten for 18F
6. the minimum acceptable shielding required for a vial is 3 mm and 3 cm of Pb for 99mTc and 18F, respectively
7. all tools increasing the distance between the hands/fingers and the source are very effective for dose reduction
8. training and education on good practices are more relevant parameters than worker's experience level itself
9. working fast is not sufficient, the use of shields or increasing the distance are more effective than pushing on the working speed.
Concerning the estimation of exposures to be received, a dose estimation tool has been developed based on the ORAMED results and it is available via the ORAMED web site (see http://www.oramed-fp7.eu/ online). This dose estimation tool provides values for the expected doses at 11 different points in each hand when preparing or injecting one the radionuclides studied within the ORAMED project.
Potential impact and main dissemination activities and exploitation of results
The project was initially defined as an applied research project with technological development. Its development has fulfilled the foreseen objectives. The composition of the consortium with representatives from research institutes, universities, hospitals, government bodies and commercial companies, as well as the coupling of experimental dose measurements and high accuracy modelling capabilities, have been very useful to reach those objectives. The main achievements shall provide impact on the following fields and target groups:
Optimisation of radiation protection of medical staff: recommendations to improve work practices in order to reduce the exposure to ionising radiation.
The measurement and simulation campaign performed in the field of IR/IC within the ORAMED project revealed a large variability of practices followed in different hospitals. It is demonstrated that, in general there are a large number of parameters that affect the extremity and eye lens doses. Medical staff in interventional radiology should follow the following recommendations to reduce their level of exposure to radiation. Only dedicated interventional equipment and rooms (properly shielded) should be used. Personal protective equipment should be used (at least collar and Pb aprons). Pb glasses with side shield should be preferred. The room protective equipment should be used and positioned properly. The ceiling suspended shield should be placed as close to the patient as possible. The combination of transparent ceiling shield and Pb drapes that touch the patient is very efficient for the protection of the eyes and hands. If biplane systems are used an extra ceiling shield to reduce the scattered radiation from the lateral tube is very important for the protection of the eyes. It is more effective if it is positioned at the side of the operator (or next to the operator). The tube should be placed below the operating table. The higher doses at the legs in this setup can be reduced by a properly positioned table shield. Care should be taken for the table shield when the operators need to move around the table for medical reasons. Mobile floor shield should be used for the assisting personnel that need to be in the irradiation room. The femoral access should be preferred whenever it is possible from the medical point of view. Going outside the operating room during the image acquisition is a practice which can reduce the doses significantly. Avoid direct exposure of hands to primary radiation. The effectiveness of the proposed recommendations can be checked by using the appropriate dosimetric systems. Monitoring of the eyes and fingers (or wrists) should be performed on routine basis. The dosimeters should be worn on the side of the operator which is closest to the x-ray tube. The dosimeter should be placed on the dorsal or palmar side of the hand when the x-ray tube is placed above or below the operating table, respectively. APDs worn above the Pb apron are recommended as operational personal dosimeter, especially for radiologists and cardiologists. Their use will increase the awareness of the personnel while they are in the operating room. Prior the selection of an active personal dosimeter, one should make sure that the device is appropriate to measure low-energy photons and pulsed radiation fields.
In NM, the ORAMED project demonstrated a large variability of measured doses in the different hospitals depending on the radiation protection means and the operator's habits. The sensitivity analysis carried out through Monte Carlo simulations employing voxel models, representing operator's hand during the considered practices was found to be very useful in order to better understand the influences of the parameters of interest to reduce the dose of the personnel. NM staff should follow the following recommendations to reduce their skin dose during the preparation (labelling) and injection of radiopharmaceuticals. Shielding of vials and syringes should be used when handling radiopharmaceuticals. The minimum acceptable thickness of shielding required for a syringe is 2 mm of tungsten for 99mTc and 5 mm of tungsten for 18F. For 90Y, 10 mm of PMMA completely shield beta radiation, but a shielding of 5 mm of tungsten provides a better protection cutting down Bremsstrahlung radiation, too. The minimum acceptable shielding required for a vial is 3 mm and 3 cm of Pb for 99mTc and 18F, respectively. For 90Y, an acceptable shielding is obtained with 10 mm of PMMA with an external layer of a few mm of Pb. Any tool increasing the distance between the hands/fingers and the source is effective for dose reduction. The use of shielding and tools to increase distance between the hands and the source are a precondition but not a guarantee for low exposures. Training and education on good practices were found to be essential. Working fast is not always sufficient. The use of shields, together with the help of tools to increase the distance to the source are, in general, more effective. ORAMED project showed that the skin effective dose limit could be exceeded while handling radiopharmaceuticals. Since the dose limit for the skin (500 mSv/year) must be applied to 'the dose averaged over any area of 1 cm² regardless of the area exposed', some recommendations on extremity monitoring in NM are given to have a good estimate of the maximum skin doses. When measuring extremity doses in NM practices, the effective thickness of the dosimeter and the position to wear it are important matters of concern. Thin detectors should be used when handling positron or beta emitters. The base of the index finger of the non-dominant hand with the sensitive part of the dosimeter placed towards the inside of the hand is the recommended position for routine extremity monitoring in NM. A rough estimate of the maximum dose to the hand can be obtained by multiplying the reading of the dosimeter worn at the base of the index finger of the non-dominant hand by a factor of six.
The abovementioned recommendations provide guidelines to medical staff to adjust their way of working in order to reduce their exposure to radiation. They are also of interest to medical physicists and radiation protection officers in order to select the best radiation protection means and to improve the personal dosimetry systematic. These recommendations were first presented in the ORAMED workshop in Barcelona in January 2011 and are now available at the project website as leaflets that can be distributed to target groups (see http://www.oramed-fp7.eu online). They should lead to a decrease of the occupational doses of medical staff.
Development of new products to better measure the ionising radiation exposure of medical staff
The collaboration between two commercial companies and research institutes, in particular national calibration laboratories, has allowed the development of two innovative prototypes to improve personal dose monitoring in interventional radiology.
In recent years an increased occurrence of radiation related lens opacities and cataracts for interventional radiologists have been reported. However, the eye lens doses are hardly ever measured in practice. In the framework of ORAMED, an eye lens TL dosimeter has been developed, optimised and tested. The dosimeter is now under the process of being patented by Radcard s.c. under the commercial name EYE-D. A sample of an EYE-D and a commercial leaflet were distributed to ORAMED 2011 workshop participants. This is the first dosimeter available commercially specially designed to provide precise measurements of radiation dose to eye lens. It can be worn close to the eye and presents a good angular and energy response in terms of Hp(3).
Up to now, APDs were not specifically designed to be used in interventional radiology typical radiation fields. The main limitations were due to the need to respond at high dose rates during short time intervals. A prototype called 'OMEGA' was developed by MIRION Technologies using a new generation of ASIC 'CID 3G' in order to replace the existing 'CID200' 1.2 µM CMOS process ASIC integrated in the DMC2000XB. Performance of the prototype is promising and it is foreseen that shortly a new product will be available.
The first improvement on the OMEGA prototype consisted in having a response that did not depend more than +/- 10% all over the energy range of interest used in IR/IC procedures. In addition the detection module was re-designed so that the dosimeter angular response was improved to a maximum deviation of +/-30% for angles up to +/-65°. Finally, the dead-time compensation correction was adjusted to present a satisfactory response up to dose rates of 10 Sv.h-1 in continuous fields and up to 20 Sv.h-1 in pulsed fields.
The new products developed in the framework of ORAMED cover two important lacks identified in the field of individual monitoring in IR/IC. The continuous feedback between manufacturers and scientists has been essential for the developments. Thus, the study has already had a first positive impact for the companies in charge of the development and can be of interest to other dosimeter manufacturers. In addition, the new devices will be useful tools for increasing awareness of medical staff on radiation exposure and for having a better assessment of personal doses in interventional radiology, in particular for the measurement of the eye lens dose. High impact in radiation protection methods is foreseen both in Europe and worldwide.
Development of new methodologies to measure eye-lens doses
Hp(3) is defined as the operational quantity to control the eye lens legal dose limits. However, in 2008, at the beginning of the project, there was no international agreement on how to reproduce this quantity in laboratory and how it could be related to basic dosimetry quantities. The theoretical investigation on the operational quantity Hp(3) undertaken in the project has provided a set of new conversion coefficients from Ka to Hp(3), defined on a new model of phantom. The proposed formalism has been essential for the development of the eye lens dosimeter and for the definition of calibration methods in this field.
International organisations, such as the International Commission on Radiation Units and Measurements (ICRU) and the International Commission on Radiation Protection (ICRP) have been contacted to inform them on the results in the study so that it can be considered in the development of new standards and international recommendations. A framework for the measurement and calibration of the operational quantity for eye lens dosimetry has been developed and this will improve the measurement and standardisation of the eye lens dose measurements for many years to come.
Development of type-tests and calibration procedures for personal dosimeters to be used in interventional radiology
A large number of international type-test and calibration standards are available for individual monitoring purposes. However, in the field of medical applications, there are several limitations that can be overcome by following some of the recommendations proposed in the framework of ORAMED.
A proposal for the type test and calibration of eye lens passive dosimeters especially used in IR/IC has been prepared. The proposal starts from the only one existing standard dealing with eye lens dosimetry using TLDs (ISO 12794) and proposes technical requirements for parameters such as, detection threshold, energy and angle responses. The recommended requirements are based on ISO 12794 and IEC 62387 and take into account the particular use of those dosimeters at IR/IC workplaces.
The proposal adopts the methodology developed in ORAMED project to reproduce and calculate in laboratory the operational quantity Hp(3). Conversion coefficients from air kerma to dose equivalent at 3 mm depth for RQR and ISO radiation qualities are presented in the proposal and the new water cylindrical phantom introduced by ORAMED is recommended for type-test and calibration. These proposals and coefficients should be taken into account by the international standard organisations when revising the relevant standards to allow a better estimate of Hlens.
The IEC 61526 standard is the international standard to be applied for type-testing of APDs. However, as pulsed radiation fields are not taken into account in existing standards, some information on the APD characteristics in pulsed field similar to those met at workplace are needed prior to using them in these fields.
The ORAMED project provides several guidelines to help the end-user in the selection and verification of the APD. Different sources of information are recommended: the results of the tests performed within the ORAMED project, specific ad-hoc tests performed by the manufacturers or some own tests performed in situ by the user.
The abovementioned proposal provides guidelines to calibration labs to perform calibration of eye-lens dosimeters and to derive the operational quantity from the reference air kerma value. Relevant recommendations should be used by medical physicists and radiation protection officers in order to select the best active personal dosimeter for their service and to make sure that their device will be appropriate for their use. The recommendations were first presented in the ORAMED workshop in Barcelona in January 2011 and are now available at the project website, as leaflets that can be distributed to target groups. Several manufacturers, present at ORAMED 2011, showed high interest on the results on APDs, both in laboratories and hospitals and encouraged the idea about having regular independent tests, as the ones performed in ORAMED.
Potential impact on regulation in radiation protection and more specifically in individual monitoring
The radiation protection recommendations and calibration guidelines are of interest to regulators and policy makers, on the one hand to disseminate good practices and on the other to up-date knowledge in radiation protection of medical staff and thus define reference levels and regulations for extremity dosimetry. Extremity dosimetry is far less developed in Europe than whole body dosimetry and there is a lack of systematic data. The situation is even worst in the case of eye-lens doses which are hardly ever measured or estimated. The results of the study highlight the importance of monitoring the wrists or fingers and eye lens.
In recent years there is growing epidemiological evidence of excess risk at the radiation doses measured within ORAMED, however, very often, dose assessment is not comprehensive in those studies. Moreover, the system of radiological protection is mainly based on excess risk of cancer induced by ionising radiation and only little information is available on non-cancer effects, such as cardio-vascular disease and lens opacities. The dose data base obtained in the ORAMED project will be of great use to contribute to new research on risks related to occupational doses. Representatives of most European radiation protection regulator bodies were present at ORAMED 2011.
Improvements in training on radiation protection for medical staff
Training material within the topics of the project has been prepared. Several proposals are available, specifically designed for different target-groups.
The material for medical staff consists of two modules (namely IR and NM) based on the experience of the project and taking advantage of images of practical situations as well as adapted to be used with interactive systems. Since there is already general training material available for trainers, we have prepared some recommendations on what training material could be used in each field and how this general information can be completed with the specific modules developed within ORAMED. IRSN has prepared some guidelines mainly based on IAEA modules available from the web site at http://rpop.iaea.org/RPOP/RPoP/Content/AdditionalResources/Training/1_TrainingMaterial/index.htm Material for dosimetry services and metrology labs was also prepared, by CEA, for dosimetry services and metrology labs, focussed on recommendations for the calibration of dosimeters to be used in interventional radiology.
SMU prepared a video for interventional radiology, showing how ORAMED measurements were performed. BfS prepared a video on Y-90 DOTA therapy. It includes good recommendations on radiation protection measures. Training material and videos are available at the ORAMED website and were presented during the ORAMED 2011 workshop.
Dissemination activities
The dissemination activities included the international workshop on 'Optimisation of radiation protection of medical staff', ORAMED 2011, which was organised from 20 to 22 January 2011, in the School of Industrial Engineering of Barcelona at UPC (Spain). Moreover, the project team participated in 10 scientific international conferences. In addition, presentations have been given to most national conferences in the topics studied in ORAMED and the final results were planned to continue to be presented during 2011 and beginning of 2012 in international conferences. Seven stages have been organised between the different groups, to let the young researchers learn from the experience of other partners and to harmonise work. ORAMED members also participated to five training activities organised by international institutions and professional societies. Finally, the ORAMED website was launched.
Project website: http://www.oramed-fp7.eu
The medical staff that carries out interventional radiology and cardiology (IR/IC) procedures is likely to receive significant radiation doses to their hands, or parts of their body not covered with protective equipment, as they are close to the x-ray field. The dose range for the same kind of procedure varies a lot, since there are many factors affecting extremity doses. In addition, there is evidence that eye lens doses can be high in IR/IC and cases of lens opacities and cataract have been reported in recent years. Besides, a lack of appropriate equipment is also identified in the field of active personal dosimeters (APDs) for typical fields in IR/IC. Very few devices can detect low energy fields and none of them are really designed for working in pulsed radiation fields. In the field of nuclear medicine (NM) the extremity doses to the medical staff are also known to be very high. One can highlight the difficulties in estimating the dose distribution across the hands and there is a need for better knowledge of doses received during the main tasks of a NM department, especially using unsealed sources.
The ORAMED project, (see http://www.oramed-fp7.eu online) was set up to optimise the working procedures in these medical fields with respect to radiation protection. The project was structured in five work packages (WPs):
1. extremity and eye lens dosimetry in IR and IC
2. development of practical eye lens dosimetry
3. optimisation of the use of APDs in IR and IC
4. extremity dosimetry in NM
5. training and dissemination.
A coordinated measurement program in European hospitals was organised both in IR/IC and NM departments. Moreover, simulations of the most representative workplaces/procedures were performed to determine the main parameters that influence the extremity and eye lens doses. Next to these, some dedicated studies on improving the eye lens dosimetry and active personal dosimetry were conducted.
Based on the measurement and simulation results, a series of practical guidelines and training packages were developed. The influence of the different radiation protection measures, such as lead (Pb) shields, has been quantified and clear monitoring requirements have been formulated for a series of medical procedures.
A formalism for the use of the operational quantities for eye lens dose measurements has been worked out (calibration phantom, conversion coefficients, type test procedures). A dedicated eye lens dosimeter has been developed during the project and is now commercially available to be used in the hospitals.
Different existing APDs have been tested for the first time under conditions that are representative of the conditions met in clinical practice (e.g. pulsed fields, low energies) and a series of guidelines for the use of these APDs in IR/IC have been made. Moreover, an improved APD device designed specifically for IR/IC fields has been developed.
The extensive measurement and simulation campaign for extremity doses in NM lead to a systematic evaluation of the different radiation protection measures. These are condensed in a series of practical guidelines to be used. The dose distribution across the hands of the technologists and the physicians was characterised and recommendations for routine monitoring have been formulated.
The outcome of the ORAMED project will lead to an optimisation of the radiation protection standards for medical staff. The practical guidelines that have been developed will be used by medical staff in hospitals for many years to come. In particular, the developments on the eye-lens dosimetry and the APDs will improve the practical measurement capabilities in the field.
Project context and objectives:
The staff that carries out interventional procedures is likely to receive significant radiation doses to their hands, or parts of their body not covered with a protective apron, as they are close to x-ray tubes. The doses for this kind of procedures vary a lot, since there are many factors affecting extremity doses such as protective devices, x-ray geometry and spectra, the scattered radiation from the patient, etc.
There are cases mentioned in the literature where the extremity doses can approach the dose limits. In these cases either the high workload or the lack of a proper radiation protection policy are responsible for the high doses observed.
Routine monitoring of extremities is difficult, since 'the most exposed area' according to the International Commission on Radiological Protection (ICRP) recommendations cannot easily be found. In most cases only finger or hand doses are reported; doses to the eye lens, legs or thyroid have not been evaluated. In some studies doses to the legs can be even higher than doses to the hands. Even when ring/wrist dosimetry is used for extremity monitoring the position of the dosimeter is not clear. There is evidence that eye lens doses are high in interventional radiology and cases of lens opacities and cataract have been reported in recent years. However, eye lens doses are never measured in routine applications and in addition very few data can be found in the literature. There is no suitable dosimeter available and the standards for the operational quantity measurements are not complete. This situation is partly due to the lack of conversion coefficients and a suitable calibration procedure.
A lack of appropriate equipment is also identified in the field of APDs for typical fields in IR/IC. Very few devices can detect low energy photon fields and none of them are really designed for working in pulsed radiation fields.
The literature concerning radiation exposure and protection of NM staff is much more limited than in the case of interventional radiology and mostly refers to conventional diagnostic NM.
As a consequence of the definition that the dose limit for the skin has to be applied to 'the dose averaged over any area of 1 cm2 regardless of the area exposed' it is advisable to measure the local skin dose at the location with presumably the highest exposure. This requirement is the central dilemma of extremity dosimetry and causes severe practical difficulties. In daily practice when preparing and administering radio-pharmaceuticals in NM it is not easy to comply with that requirement since it is often not known which part of the hand receives the highest dose. Moreover, the dose distribution over the hand may vary during a single process as well as when various workers perform the same procedure.
Unsealed radiation sources are being increasingly used in NM for radiation therapy, in particular, nuclides that emit beta or mixed beta/gamma radiation. Considering the preferential use of beta emitters, the dosimeters must be appropriate for beta radiation, taking into account both the energy spectra of the nuclides and the spectral dose response of the dosimeter. This means that in NM therapy, staff may be exposed to high doses, even exceeding the annual limit of the dose to the skin of 500 mSv. Thus, adequate safety measures including an accurate individual monitoring of the personnel are a strict requirement.
It should be pointed out that most general official personal monitoring databases, such as the United Nations Scientific Committee on the Effects of Atomic radiation (Unscear) or the Information System on Occupational Exposure (ISOE), do not include extremity doses and that many countries with NM services do not have authorised ring dosimetry services. Furthermore, the data available in official national registers are much lower than the estimated doses in research studies, which probably means that monitoring is not being done correctly. Through the dissemination of the ORAMED results and the training, we will aim at the improvement of the radiation protection in these fields.
The ORAMED project proposes to develop methodologies for better assessing and reducing exposures to medical staff. This general objective is being achieved through the development of five main topics, structured in five WPs.
1. The objective of WP1 was to obtain a set of standardised data on doses for staff in IR/IC and to optimise staff protection.
2. The objective of WP2 was to establish a sound theoretical and experimental basis to assess eye lens doses.
3. The objective of WP3 was to optimise the use of APDs in interventional radiology.
4. The objective of WP4 was to detect the most exposed part of the skin by measuring the extremity doses and dose distributions across the hands of the medical staff working in NM departments.
5. The last objective of the project (WP5) was to design and develop an accurate teaching and knowledge dissemination program and to make sure that the conclusions and recommendations of the project are transmitted to the stakeholders, mainly medical staff, radiation protection officers, dosimetry services and authorities in the field.
Project results:
WP1: Extremity dosimetry and eye lens dosimetry in IR/IC
Measurement campaign
List of procedures
The list of procedures includes three cardiac and five general interventional diagnostic and therapeutic examinations. More specifically, the list is composed of cardiac angiographies (CA) and angioplasties (PTCA), radiofrequency ablations (RFA), pacemaker and cardiac defibrillator implantations (PM/CD), angiographies (DSA) and angioplasties (PTA) of the lower limbs (LL), the carotids and the brain (C/B) and the reins (R), embolisations and endoscopic retrograde cholangiopancreatographies (ERCP). The choice of the procedures was based on their potential impact on the annual exposure of the staff. Thus two main selection criteria were defined: high annual frequency and possible high kerma area product (KAP) values. However, some procedures of low frequency were also considered in the study, in order to include the different parts of the patient body that is irradiated, or because they are often performed in rooms with limited shielding equipment.
Measurement protocol
A measurement protocol was established, in order to obtain an accurate description of the staff irradiation scenario and to homogenise the collected data. Thus, different parameters related to the angiographic system, the type and complexity of the procedure, the position of the physician, the protective equipment, the experience of the physician, some field parameters and finally the fluoroscopy time, the number of images acquired and the KAP values were recorded.
For the measurements it was decided to use high sensitivity thermoluminescent dosimeters (TLDs). The TLDs were sealed in small plastic bags and taped on the parts of the body to be monitored.
Results
Among the types of cardiac procedures that were included in the measurement campaign, the doses to the operators are higher during the pacemakers and implantation of cardiac defibrillators (PM/ICD), even though the respective KAP values are relatively low since only fluoroscopy is used. During these procedures the operators work very close to the irradiation field and most of the time without any protective shield.
Among the IR procedures special attention should be given to embolisations, particularly to the doses to the eye lenses. Operators are also significantly exposed during therapeutic procedures such as angioplasties of the lower limbs and the renal arteries. During cerebral and carotid procedures the doses are relatively low since femoral access is usually used and the operator stands at a larger distance from the irradiated part of the patient's body compared to other procedures performed in the thoracic or abdominal region.
Finally, for ERCP procedures the doses are generally low. Special care should be taken regarding the use of a ceiling suspended shield, especially for the protection of the eyes, when overcouch irradiation is used.
The position of the maximum dose was found to be at the part of the operator which is close to the x-ray tube. More generally, in all the procedures studied the left wrist was found to be the position where the maximum dose was recorded most often, followed by the left finger. The maximum was found generally at the left wrist for femoral access and at the left finger for radial or direct access, the operator being then closer to the beam field. Therefore, both wrist and ring dosimeters are suitable for the routine monitoring. However, taking into account the respective annual limits for all positions the eye lens exposure is more important.
According to the information on the annual workload that was gathered from 84 physicians, the annual limit for the skin on the fingers could be exceeded only for a few workers but the annual doses were estimated based only on the procedures for which the operator was monitored during the measurement campaign.
In the case of the CA/PTCA procedures, 5 to 15 % of the workers seem possible to receive an annual dose above the three tenths of the annual limit. Furthermore, there are around 5 % of the workers that could even exceed the annual skin effective dose of 500 mSv for PM/ICD procedures. Thus, the monitoring of the hands is definitely a requirement for interventional procedures especially in interventional cardiology. The annual doses were considered low for the ERCP procedures, so the monitoring of the hands is not necessary in this case.
Although the annual doses to the legs can be high for certain kind of procedures, it was shown within the analysis performed in WP1 that this was mainly happening for tube, below couch configurations without any shield attached to the operating table. Therefore, the monitoring of the doses to the legs is not needed when a table shield is properly used.
Finally, the eye lens doses can be high so monitoring is also recommended for all investigated procedures, except ERCP. Especially, if the dose limit for the eye is decreased in the future, monitoring and protection measures will be even more important.
Simulations
Description of the input file
The numerical simulations have been performed using the Monte Carlo n-particle extended (MCNP)-X code. The medical internal radiation dose (MIRD) type anthropomorphic models have been used for the simulation of the patient and the operator. The 'patient' phantom is at supine position and the 'operator' one is standing close to it, in a configuration that is typical in an IR and IC procedure. The original model was modified in order to represent more realistically the irradiation scenario: eyes have been added, the arms have been redefined and forearms and hands are bent in a more realistic position. A thyroid collar of 0.5 mm Pb and an apron of 0.5 mm Pb in front of the body have also been added. Finally, a cell filled with air representing the KAP chamber and the image intensifier have been added to the input file. F6 tallies were used for the calculation of the doses to the eye lenses, hands, wrists and legs. The tallies were positioned at 0.07 mm and at 3 mm depth for the calculation of Hp(0.07) and Hp(3) respectively.
Geometry characteristics
The x-ray tube was simplified to a point source emitting a photon beam in a cone with an aperture corresponding to that desired for the simulation of the x-ray field diameter. The energy spectra for the selected kVp and filtrations were determined using the x-ray data of the Institute of Physics and Engineering in Medicine (IPEM), report 78. Moreover, the Pb collimator was not simulated explicitly, but defined as a volume killing all the photons entering inside it.
The first parameters examined within the simulation campaign are the tube voltage and the filtration. More specifically, the tube voltage was changed from 60 to 110 kVp and the filtrations from 3 to 6 mm aluminium (Al) and from 0 to 0.9 mm copper (Cu). Moreover, simulations also were performed to examine several cases of beam projections, field size, different positions of the operator, protective glasses for the operator, protective equipment for the room, and four different parts of the body being irradiated, namely head and neck, thorax, abdomen/pelvis and lower limbs.
Results
The beam projections for which the operator is the most exposed are the left anterior oblique (LAO), where the operator stands at the side of the X-ray tube and the cranial projections, where the image intensifier is towards the patient's head. When comparing the overcouch (AP) with undercouch (PA) irradiations, the doses are found significantly lower for the undercouch one, up to 12 times for the eyes, eight times for the hands and four times to the wrists, for the selected geometries. In case of PA the dose to the legs can be relevant if no table shield is used. For the lateral projections, the right anterior oblique (RAO) should be preferred, as the doses were found lower compared to the LAO, up to three times for the eyes, three times for the hands and 22 times for the wrists, for the selected geometries.
The table shield is very effective for the protection of the legs and reduces the dose from 83 % to 99 % for the geometries and setups that were tested. The Pb glasses, for the specific geometry that was examined, are very effective for the protection of the eye lens which is closer to the x-ray tube. In practice, Pb glasses that cover well the eyes and also have lateral protection are advised. Lens thickness of more than 0.5 mm Pb doesn't improve the protection of the eye lenses significantly.
A ceiling suspended shield is also very effective for the protection of the eye lenses; its use is essential especially when Pb glasses are not available or cannot be worn for practical reasons and when overcouch irradiation is used. This type of shield also provides protection to the hands and wrists. The simulations showed how important it is to correctly place the ceiling shield. Even a small gap between the patient and the shield reduces the effectiveness of the shield to the hands. For this reason a shield with Pb stripes attached at the bottom so that it touches the patient is advised. For the lateral projections the eyes are better protected when the shield is positioned closer to the side of the operator and not just above the patient. In practice, a second shield is advised especially when biplane systems are used, so that the operator is protected from both x-ray tubes. For all monitored positions the doses are higher when larger field sizes are used.
Concerning the beam quality, more energetic beams lead to lower doses to the operator up to 60 %. However, one should keep in mind the deterioration of image quality and contrast in this case.
The measurement and simulation campaign performed within the ORAMED project revealed a large variability of practices followed in different hospitals. As a consequence, the measured doses, even for specific procedures, vary significantly from one case to another. On the other hand, the simulation data showed the way that each parameter separately can influence the extremity and eye lens doses and not in combinations as it is the case from the measurement results. The combined data led to the following recommendations. It is important to note that some of the proposed guidelines cannot easily be adopted since there are restrictions from the medical point of view. However, some of them are easily adjustable and can improve the protection of the medical staff significantly:
1. the equipment used for IR/IC should fulfil specific requirements and standardisation in their design, manufacture, acceptance and maintenance
2. personal protective equipment should be used for all the personnel in the room
3. the ceiling suspended shield should be placed just above the patient, especially in the cases that the tube is above the operating table; the operator should stand well behind it
4. when ceiling suspended shield is not available or cannot be used for practical reasons protective Pb glasses should be used
5. the table shield should be always properly adjusted to protect both legs
6. the tube should be placed below the operating table when C-arm systems are used. However, the increase at the leg doses for this setup has to be compensated by the use of a properly positioned table shield.
7. if biplane systems are used, the proper use of lateral shield is very important for the protection of the eyes
8. mobile floor shield should be used for the assisting personnel that need to be in the irradiation room
9. the femoral access of the catheter should be preferred, if it is possible from the medical point of view, compared to the radial one
10. the use of an automatic image injector which allows the operator to leave the room during the image acquisitions is a practice which can reduce the doses significantly, especially to the hands
11. the operators should avoid direct exposure of hands to primary radiation
12. monitoring of the eyes and fingers should be performed on routine basis. The dosimeters should be worn on the side of the operator which is closest to the x-ray tube. The finger (or wrist) dosimeter should be placed on the dorsal or palmar side of the hand when the x-ray tube is placed above or below the operating table, respectively.
WP2: Development of practical eye lens dosimetry
The basic step of WP2 was to revise the approach for the definition and calculation of conversion coefficients for Hp(3). Relying on morphological studies on the head, it was decided to propose a cylindrical (20 cm diameter and 20 cm height) simplified four-element tissue equivalent phantom on which to calculate the operational quantity.
The discussion was especially motivated by the fact that the possible adoption of the trunk phantom also for calculating Hp(3) seemed to be rather inappropriate due to its strong deviation from the reality in terms of mass and shape. In particular the 30 x 30 cm2 plane face is too large and of different shape than the head. Moreover, the edges play a very important role when looking at the angular dependence of the operational quantity. Such effects also influence the calibration procedure, if such phantom is adopted in the procedure, as its backscatter characteristics are not corresponding to what happens in practice to a target organ like the eye embedded in the head.
A new set of photon conversion coefficients were obtained. The calculations demonstrated the better suitability of the adoption of the cylinder to reproduce the organ equivalent dose (eye lens) in various irradiation conditions.
It has to be pointed out that, as expected the operational quantity as determined on the proposed cylinder are in satisfactory agreement with the limiting quantity. Moreover it should be noted that for nearly grazing incidence that could occur in IR/IC procedures, in which the patient acts as a scattering object, the operational quantity as determined in the trunk slab is not conservative estimate of HT(eye lens). The present general considerations justify the adoption of the new phantom to develop a self-consistent eye lens dosimetry procedure. It has to be pointed out that the new calculated operational quantity Hp(3) based on the cylinder is also better fitting with the HT(eye lens) values actually reported in ICRP-74 and ICRU-57.
RADCARD dosimeter prototype
On the basis of the studies on Hp(3) and the new set of conversion coefficients produced, a prototype was developed by RADCARD with the support of ENEA and CEA. The dosimeter had to measure correctly Hp(3) for eye-lens while being comfortable for users and for dosimetric services, waterproof and inexpensive. The characteristics to be optimised were the energy response and the angular response. The study was performed mainly varying the TLD detector type and dimension, the capsule material and dimension. The experimental irradiations were carried out using beams from the RQR and narrow series. The first series, with a moderate filtrationi better reproduce the operative condition met in IR/IC, whilst the second, with high filtration is closer to mono-energetic photons.
The prototype was finally tested at the CEA-LNE Laboratories in Saclay (France) and the irradiation results were compared with Monte Carlo simulations. The calculations were performed with the narrow spectrum series. The results are in a very satisfactory agreement.
Establishing a Protocol for type test and calibration of personal dosimeters in terms of Hp(3)
This part of the work was aimed at making a proposal for the type test and calibration of eye lens passive dosimeters especially designed to be used in IR/IC. Starting from the only one existing standard dealing with eye lens dosimetry using TLDs parameters such as, detection threshold, energy and angle responses have been reviewed and have been harmonised, as much as possible, with the International Electrotechnical Commission (IEC) 62387 requirements, taking into account the particular use at IC/IR workplaces. Conversion coefficients from air kerma to dose equivalent at 3 mm depth for RQR and International Organisation for Standardisation (ISO) radiation qualities, employed for type test and calibration purposes, have been calculated in the new phantom introduced by ORAMED.
Type tests are intended to demonstrate that the dosimeters are suitable for measurements in workplace conditions. For dosimetry systems based on passive personal dosimeters, two international standards covering photon and beta radiations exist for type testing: IEC 62387-1 and ISO 12794. The comprehensive document treats the following aspects:
1. wearing conditions at the IR/IC workplaces and consequences on type test criteria
2. detection threshold
3. performance requirements for energy response
4. performance requirements for angle response
5. other requirements
6. Calibration.
Conclusions
WP2 of the ORAMED project allowed establishing a new and self-consistent approach for an improved dosimetry procedure on the eye lens. Starting from the proposal of a well suited simple phantom for the theoretical quantity, the corresponding designed calibration phantom was fabricated at a very low cost. Thereafter the phantom was adopted during the ORAMED project for calibrations of the prototype. It was developed to respond in the best way in terms of Hp(3), according to the protocol that was conceived to offer clear guidelines specifically addressed to the eye lens personal dosimetry. The obtained results can be of help to the dosimetry community working in the medical field and also in other disciplines.
WP3: Optimisation of the use of APDs in IR/IC
Tests of eight commercial APDs in laboratory conditions with continuous and pulsed x-ray beams and in hospitals
The tests with continuous x-ray fields were made in two calibration laboratories. These tests were performed to determine the response of selected APDs in terms of personal dose equivalent, energy, personal dose equivalent rate and angle. The reference fields were used as defined in the ISO 4037-1 standard. Three measurements per APD were performed. Two dosimeters of each type were tested. Dosimeters were placed on an ISO slab phantom. The results were analysed considering the requirements of the IEC 61526 standard.
Results
The personal dose equivalent response of the tested APDs is linear in the dose range of interest in radiation protection, i.e. up to 500 mSv. The energy response of the tested APDs is within the interval as required in the IEC 61526 standard, from S-Co energy down to N-30 for all APDs except EDD30 and DoseAware. EDD30 energy response is within the IEC requirements between N-80 and N-20 qualities; dose Aware energy response is within requirements between N-120 and N-40. Most APDs can stand high dose rates up to 10 Sv.h-1 except PM1621A, for which the response diverges rapidly from 1 Sv.h-1 as well as EDD30 and DoseAware which saturate for personal dose equivalent rates above 2 and 4 Sv.h-1 respectively. It is interesting to notice that most APDs can stand personal dose equivalent rates higher than those indicated in their technical note. The angular response is within the interval for energies down to N-30 for all APDs, apart from AT 3509C for which the angular response is inside the before mentioned interval at 60 ° only for N-80.
Tests of APDs with pulsed x-ray beams
The tests in pulsed mode were performed at the French standard laboratory for ionising radiation. The influence of several parameters in different conditions on the response of the APD in pulsed mode was studied.
For most APDs, the response decreases when the personal dose equivalent rate increases. For personal dose equivalent rates lower than 2 Sv.h-1 the responses are, in general, close to one and fall down more or less rapidly for higher dose rates, except DIS-100 that gives a correct response up to 55 Sv.h-1. It was noticed that PM1621A, equipped with a Geiger-Muller tube, does not provide any signal at all in pulsed mode. DMC 2000XB, EPD Mk2.3 EDMIII, EDD30, AT3509C and DoseAware contain all silicon detectors. The difference of their response to the pulsed mode is probably due to the time response of their electronic systems. DIS-100, which has a 'hybrid' technology between silicon and ionisation chamber, presents correct results.
The variation of the APD response with the pulse frequency between 1 and 20 s-1 is roughly equal to 10 % for EDMIII, EDD30 and DoseAware and to 30 % for the other devices (except PM1621A).
When the pulse width is larger than 1 s, the responses in pulsed and continuous radiation field are quite similar. No significant effect of pulse width was observed on the response of all APDs. All results from the pulsed field tests show that the longer the pulses and the higher the frequency, the better the behaviour of the devices tends to be.
The first series of tests in real conditions were performed by positioning APDs on an ISO slab phantom representing the operator. The scattered irradiation was produced by an anthropomorphic Rando-Alderson phantom representing the patient. The tests were performed on an x-ray system (PHILIPS BZR79 Optimus). The APDs tested were: DMC 2000XB (MGPi), EPD Mk2.3 (Siemens), EDM III (Dosilab), PM1621A (Polimaster), DIS-100 (Rados) and EDD30 (Unfors).
As reference the routine passive TLD from the Belgian Nuclear Research Centre was used (uncertainty 20 %). Both the APD as the TLD were positioned together on the ISO slab phantom. The dose uniformity on the surface of the phantom was within 20 %. The thorax of the RA-phantom was irradiated and different realistic set-ups were considered. The main objective of these tests was the study of the behaviour of the APDs in realistic conditions with the possibility to select specific field parameters.
In a range of dose equivalent rates tested from 10 mSv.h-1 to 1.8 Sv.h-1 the APD response is within 50 % for the range of dose equivalent rates tested, except for the EDM III for which the dose is general higher than the TLD dose and except PM1621A that does not give any signal, which is consistent with the laboratory tests in pulsed fields. No important influence of the tube voltage and of the pulse width was observed on the APD response compared to the TLD.
Tests on operators
For these series of tests operators wore, side by side, one APD and one passive dosimeter above the Pb apron. The dosimeters were worn during several interventions to integrate doses of at least 300 µSv for several types of IR/IC procedures. The dose equivalent was provided by the passive dosimeter according to the routine measurement protocol of the respective partner that performed the measurements. For practical reasons only 5 APD types were tested: DMC 2000XB (MGPi), EPD Mk2.3 (Siemens), EDM III (Dosilab), DIS-100 (Rados) and DoseAware (Philips). In total 102 measurements were performed in seven different European hospitals. The main objective was to compare the measurements performed by the APD and passive dosimeter worn in routine practice where all kinds of procedures and parameter settings are used and without an accurate knowledge of these parameters.
With respect to passive dosimeters, in general all five tested APDs under-responded. We can observe a large spread in the results, which might be explained by non-uniform irradiations or the shielding of one dosimeter by the other.
Some recommendations were prepared within the group to help in selection and use of APDs at IR/IC workplaces. These recommendations are compiled on a three page leaflet available on the ORAMED website at http://www.oramed-fp7.eu/ They concern the selection of an APD in IR/IC and the use of an APD in IR/IC.
In addition, a prototype of an improved APD for IR was developed by MGPInstruments. The objective was to propose technical solutions to improve the response of an APD for an application in IR/IC based on the results of the tests in laboratory conditions. The different technical solutions developed by MIRION Technologies during this period were tested in laboratories. A prototype called 'OMEGA' using a new generation of ASIC 'CID 3G' was developed in order to replace the existing 'CID200' 1.2 µM CMOS process ASIC integrated in the DMC2000XB.
The first improvement on the OMEGA prototype consisted in having a good energy response. The second task was to reduce the angle influence on the dosimeter response. The position of the two detectors and the shape of the detection module have been re-designed in order to have a good isotropy for angles of irradiation greater than +/- 60°. In addition, a study of the dead time compensation was conducted in order to reach dose rates as high as 10 Sv.h-1.
WP4: Extremity dosimetry in NM
The objectives set out in WP4 were to evaluate extremity doses and dose distributions across the hands, to study the influence of protective devices and to propose 'levels of indicative reference doses' for each standard NM procedure, with the final goal of reducing, when possible, hand dose levels. One of the main tasks was to perform an extensive measurement campaign of hand doses in European hospitals using a unified measurement protocol.
The preparation and administration of both diagnostic and therapy procedures were studied. In diagnostics, radiopharmaceuticals labelled with 99 mTc and 18F were included in the investigations because of their wide use. For therapeutic procedures the studies were focussed on 90Y labelled radiopharmaceuticals.
For each measurement, preparation and administration to the patient of the radiopharmaceutical were separated. Eleven TLDs, calibrated to measure the personal dose equivalent Hp(0.07) were taped on gloves at 11 positions of both left and right operator's hands. For practical reasons in therapy a given pair of gloves was used for a single preparation. For diagnostics the gloves were used several times by the same operator in order to integrate significant doses in the TLDs. The identification of the hospital and operator were recorded, together with the manipulated activities, shield(s) used and any difficulty that could occur during the operations. For most of the operators the measurement was repeated five times. In diagnostics, those workers with less than four set of measurements were rejected. The TLDs used by the partners were of different types. An intercomparison exercise was done for reference to ensure consistency of the results between different partners and to select specific techniques for specific applications.
The statistical analysis was performed with 641 measurements, collected in around 20 NM departments per procedure from six European countries. For each radionuclide and procedure the analysis was performed independently. The normalised Hp(0.07) (µSv/GBq) measured at each of the 22 monitored positions were averaged over the series of measurements for each worker. These mean values were used for the analysis. The maximum normalised local skin dose was calculated as the highest of those 22 mean values.
Many parameters and steps affect the local skin dose at hands, especially for preparation. In addition there was a lack of information on potential parameters of influence, such as the operation time. Furthermore, the tools to reduce finger doses were sometimes used in some steps and not in others during a single measurement, or were either used differently from one measurement to another. The number of times that the activity was manipulated was not taken into account. In general, the fact that the measurements were not systematically watched or recorded on video could cause some important details to be missing. The problem is thus complex and therefore the analysis had to be kept simple. In spite of this, the study provides a good overview of the level of finger exposure from a wide range of working habits and working procedures.
Workers were classified according to their maximum dose, in increasing order. Some workers were considered as outliers. Those workers have practices that significantly differed from those of the majority, leading to very high or very low exposures. It has been observed that most of the workers receiving very high doses did not use shielding for vials and/or syringes. On the other hand, those with very low exposures used semi-automatic equipment, which is not a common practice yet.
Very large ranges of maximum doses were found for the same procedure. The preparation of radiopharmaceuticals involves higher finger doses per activity than for the administration, because the procedures are longer and there are more steps requiring manipulations of the vials/syringes with higher activities, some of them without shield. The dose distribution across the hands was also studied. For all procedures it was observed that the non-dominant hand usually receives higher doses than the dominant hand. For all diagnostics and therapy procedures, the index tip of the non-dominant hand is the position where the maximum dose is most frequently received, followed by the thumb of the same hand for almost all procedures. Less frequently, the same positions of the dominant hand were also found to be common positions with maximum dose. In the majority of the therapy cases, as for diagnostics, the tip of the index finger or the thumb on the non-dominant hand was found to receive the maximum dose, especially in those cases where the exposure was high.
In order to identify the parameters of influence on the skin dose, workers were classified into categories for those parameters for which information was available and whenever the amount of data in each category was large enough. For all these cases the Mann Whitney-U test was applied to analyse the differences between the skin doses received by workers within each of the categories. The results of the test did not show statistically significant differences between the doses received by experienced and beginner workers. On the other hand, the shield was found to be a very important parameter of influence both for the vial and for the syringe, causing the differences between the doses received when using shield and when not using it to be statistically significant. When feedback was given to workers after a measurement series, informing them of their exposures and with a discussion of bad practices in particular cases, a decreased dose was observed in the subsequent measurements. That is of particular importance for radiation protection optimisation because it demonstrates that the correction of the bad habits, that usually requires a small effort, could imply a very large spare in extremity doses to the operator, e.g. by using shielding and tools to avoid any direct contact of the fingers with the source.
Wrist or ring dosimeters are typically used for routine monitoring. Although there is not a harmonised criterion for the position of the ring dosimeter, in practice it is usually placed at the base of the index, middle or ring fingers of the dominant hand since these positions do not hamper work causing an underestimation of the maximum dose. The underestimation was assessed, in a first step, by calculating the correlations between the dose at all measuring positions and the maximum dose. Although the routine monitoring positions do not correspond to the position of the maximum skin dose, they can be used to estimate this quantity.
In a second step the ratios between the maximum dose and the dose at relevant monitoring positions and at the index tip were calculated. The calculation was made for each single measurement and then averaged over the set of measurements of each worker and for all diagnostics procedures separately, but the differences among procedures were not found to be relevant. Due to this reason and considering that NM workers are usually involved in more than one diagnostic procedure, the ratios were also calculated by including all data from all diagnostics procedures.
The mean ratios are significantly higher for the wrist positions. The lowest mean ratios were found for the index tip position. The ratios are also lower for the base of the index finger than for the base of the ring finger and lower for the non-dominant hand than for the dominant one. Thus, according to these results, the use of wrist dosimeters should be avoided because of a very high underestimation and a lower correlation to the maximum dose.
The annual dose of the monitored workers involved in diagnostic procedures for the ORAMED project has been estimated. For this estimation, only those procedures from which measured values, from the ORAMED measurement campaign, where available for a specific worker have been considered. Their workload and the activity manipulated per year for each radionuclide, was considered. Tthe estimated annual dose is above 150 mSv, i.e. three tenths of the annual limit, for 51 % of the workers and for 20 % of the workers the annual dose limit was exceeded.
It has to be noticed that the real situation is more complex since usually a given worker will not perform only one but several different procedures. Therefore, the annual maximum dose is likely to be an underestimation whenever the worker is actually involved in more procedures than those actually measured. The results of the estimation highlight the need to monitor the worker and to optimise the radiation protection standard in NM.
Monte Carlo simulations were employed to determine at what extent the range of doses evaluated during measurements could be considered 'intrinsically related' to those procedures and in which way it was possible to estimate the effectiveness of different adopted methodology and shielding, with such large variability of data. More than 200 simulations were performed.
Voxel models representing the hands of the operator during some selected steps of administration and preparation procedures were obtained through the elaboration of computed tomography (CT) scans of real phantoms. The radioactive source was simulated as a cylinder of proper volume representing the manipulated syringe or the vials. A sensitivity analysis was performed.
Simulations show that, depending on the radionuclides and on the voxel hand model, factors ranging from of 0.5 to 3 can be reached only by considering small displacements of the source with respect to the sensor positions. These fluctuations are of the order of those encountered in the analysis of the measurements results. Moreover, they showed the advantages concerning dose reduction of a correct use of the shielding during preparation of radiopharmaceuticals and of the convenience of employing forceps as an additional protective factor also in case of shielded sources. Concerning the shielding, large reduction factors can be obtained when using the appropriate shielding. With Monte Carlo simulations and dose mapping it was also possible to investigate at what extent the positions initially considered for the TLDs in the measurement program were suitable to estimate the maximum of the dose.
The following recommendations were derived from the observations and results of the WP4 of the ORAMED project:
1. extremity monitoring is a necessity in NM
2. the base of the index finger of the non-dominant hand with the sensitive part of the dosimeter placed towards the inside of the hand is the recommended position for routine extremity monitoring in NM
3. an estimate of the maximum dose to the hands can be obtained by multiplying the reading of the dosimeter worn at the base of the index finger of the non-dominant hand by a factor of six
4. shielding of vials and syringes is essential
5. the minimum acceptable thickness of shielding required for a syringe is 2 mm of tungsten for 99mTc and 5 mm of tungsten for 18F
6. the minimum acceptable shielding required for a vial is 3 mm and 3 cm of Pb for 99mTc and 18F, respectively
7. all tools increasing the distance between the hands/fingers and the source are very effective for dose reduction
8. training and education on good practices are more relevant parameters than worker's experience level itself
9. working fast is not sufficient, the use of shields or increasing the distance are more effective than pushing on the working speed.
Concerning the estimation of exposures to be received, a dose estimation tool has been developed based on the ORAMED results and it is available via the ORAMED web site (see http://www.oramed-fp7.eu/ online). This dose estimation tool provides values for the expected doses at 11 different points in each hand when preparing or injecting one the radionuclides studied within the ORAMED project.
Potential impact and main dissemination activities and exploitation of results
The project was initially defined as an applied research project with technological development. Its development has fulfilled the foreseen objectives. The composition of the consortium with representatives from research institutes, universities, hospitals, government bodies and commercial companies, as well as the coupling of experimental dose measurements and high accuracy modelling capabilities, have been very useful to reach those objectives. The main achievements shall provide impact on the following fields and target groups:
Optimisation of radiation protection of medical staff: recommendations to improve work practices in order to reduce the exposure to ionising radiation.
The measurement and simulation campaign performed in the field of IR/IC within the ORAMED project revealed a large variability of practices followed in different hospitals. It is demonstrated that, in general there are a large number of parameters that affect the extremity and eye lens doses. Medical staff in interventional radiology should follow the following recommendations to reduce their level of exposure to radiation. Only dedicated interventional equipment and rooms (properly shielded) should be used. Personal protective equipment should be used (at least collar and Pb aprons). Pb glasses with side shield should be preferred. The room protective equipment should be used and positioned properly. The ceiling suspended shield should be placed as close to the patient as possible. The combination of transparent ceiling shield and Pb drapes that touch the patient is very efficient for the protection of the eyes and hands. If biplane systems are used an extra ceiling shield to reduce the scattered radiation from the lateral tube is very important for the protection of the eyes. It is more effective if it is positioned at the side of the operator (or next to the operator). The tube should be placed below the operating table. The higher doses at the legs in this setup can be reduced by a properly positioned table shield. Care should be taken for the table shield when the operators need to move around the table for medical reasons. Mobile floor shield should be used for the assisting personnel that need to be in the irradiation room. The femoral access should be preferred whenever it is possible from the medical point of view. Going outside the operating room during the image acquisition is a practice which can reduce the doses significantly. Avoid direct exposure of hands to primary radiation. The effectiveness of the proposed recommendations can be checked by using the appropriate dosimetric systems. Monitoring of the eyes and fingers (or wrists) should be performed on routine basis. The dosimeters should be worn on the side of the operator which is closest to the x-ray tube. The dosimeter should be placed on the dorsal or palmar side of the hand when the x-ray tube is placed above or below the operating table, respectively. APDs worn above the Pb apron are recommended as operational personal dosimeter, especially for radiologists and cardiologists. Their use will increase the awareness of the personnel while they are in the operating room. Prior the selection of an active personal dosimeter, one should make sure that the device is appropriate to measure low-energy photons and pulsed radiation fields.
In NM, the ORAMED project demonstrated a large variability of measured doses in the different hospitals depending on the radiation protection means and the operator's habits. The sensitivity analysis carried out through Monte Carlo simulations employing voxel models, representing operator's hand during the considered practices was found to be very useful in order to better understand the influences of the parameters of interest to reduce the dose of the personnel. NM staff should follow the following recommendations to reduce their skin dose during the preparation (labelling) and injection of radiopharmaceuticals. Shielding of vials and syringes should be used when handling radiopharmaceuticals. The minimum acceptable thickness of shielding required for a syringe is 2 mm of tungsten for 99mTc and 5 mm of tungsten for 18F. For 90Y, 10 mm of PMMA completely shield beta radiation, but a shielding of 5 mm of tungsten provides a better protection cutting down Bremsstrahlung radiation, too. The minimum acceptable shielding required for a vial is 3 mm and 3 cm of Pb for 99mTc and 18F, respectively. For 90Y, an acceptable shielding is obtained with 10 mm of PMMA with an external layer of a few mm of Pb. Any tool increasing the distance between the hands/fingers and the source is effective for dose reduction. The use of shielding and tools to increase distance between the hands and the source are a precondition but not a guarantee for low exposures. Training and education on good practices were found to be essential. Working fast is not always sufficient. The use of shields, together with the help of tools to increase the distance to the source are, in general, more effective. ORAMED project showed that the skin effective dose limit could be exceeded while handling radiopharmaceuticals. Since the dose limit for the skin (500 mSv/year) must be applied to 'the dose averaged over any area of 1 cm² regardless of the area exposed', some recommendations on extremity monitoring in NM are given to have a good estimate of the maximum skin doses. When measuring extremity doses in NM practices, the effective thickness of the dosimeter and the position to wear it are important matters of concern. Thin detectors should be used when handling positron or beta emitters. The base of the index finger of the non-dominant hand with the sensitive part of the dosimeter placed towards the inside of the hand is the recommended position for routine extremity monitoring in NM. A rough estimate of the maximum dose to the hand can be obtained by multiplying the reading of the dosimeter worn at the base of the index finger of the non-dominant hand by a factor of six.
The abovementioned recommendations provide guidelines to medical staff to adjust their way of working in order to reduce their exposure to radiation. They are also of interest to medical physicists and radiation protection officers in order to select the best radiation protection means and to improve the personal dosimetry systematic. These recommendations were first presented in the ORAMED workshop in Barcelona in January 2011 and are now available at the project website as leaflets that can be distributed to target groups (see http://www.oramed-fp7.eu online). They should lead to a decrease of the occupational doses of medical staff.
Development of new products to better measure the ionising radiation exposure of medical staff
The collaboration between two commercial companies and research institutes, in particular national calibration laboratories, has allowed the development of two innovative prototypes to improve personal dose monitoring in interventional radiology.
In recent years an increased occurrence of radiation related lens opacities and cataracts for interventional radiologists have been reported. However, the eye lens doses are hardly ever measured in practice. In the framework of ORAMED, an eye lens TL dosimeter has been developed, optimised and tested. The dosimeter is now under the process of being patented by Radcard s.c. under the commercial name EYE-D. A sample of an EYE-D and a commercial leaflet were distributed to ORAMED 2011 workshop participants. This is the first dosimeter available commercially specially designed to provide precise measurements of radiation dose to eye lens. It can be worn close to the eye and presents a good angular and energy response in terms of Hp(3).
Up to now, APDs were not specifically designed to be used in interventional radiology typical radiation fields. The main limitations were due to the need to respond at high dose rates during short time intervals. A prototype called 'OMEGA' was developed by MIRION Technologies using a new generation of ASIC 'CID 3G' in order to replace the existing 'CID200' 1.2 µM CMOS process ASIC integrated in the DMC2000XB. Performance of the prototype is promising and it is foreseen that shortly a new product will be available.
The first improvement on the OMEGA prototype consisted in having a response that did not depend more than +/- 10% all over the energy range of interest used in IR/IC procedures. In addition the detection module was re-designed so that the dosimeter angular response was improved to a maximum deviation of +/-30% for angles up to +/-65°. Finally, the dead-time compensation correction was adjusted to present a satisfactory response up to dose rates of 10 Sv.h-1 in continuous fields and up to 20 Sv.h-1 in pulsed fields.
The new products developed in the framework of ORAMED cover two important lacks identified in the field of individual monitoring in IR/IC. The continuous feedback between manufacturers and scientists has been essential for the developments. Thus, the study has already had a first positive impact for the companies in charge of the development and can be of interest to other dosimeter manufacturers. In addition, the new devices will be useful tools for increasing awareness of medical staff on radiation exposure and for having a better assessment of personal doses in interventional radiology, in particular for the measurement of the eye lens dose. High impact in radiation protection methods is foreseen both in Europe and worldwide.
Development of new methodologies to measure eye-lens doses
Hp(3) is defined as the operational quantity to control the eye lens legal dose limits. However, in 2008, at the beginning of the project, there was no international agreement on how to reproduce this quantity in laboratory and how it could be related to basic dosimetry quantities. The theoretical investigation on the operational quantity Hp(3) undertaken in the project has provided a set of new conversion coefficients from Ka to Hp(3), defined on a new model of phantom. The proposed formalism has been essential for the development of the eye lens dosimeter and for the definition of calibration methods in this field.
International organisations, such as the International Commission on Radiation Units and Measurements (ICRU) and the International Commission on Radiation Protection (ICRP) have been contacted to inform them on the results in the study so that it can be considered in the development of new standards and international recommendations. A framework for the measurement and calibration of the operational quantity for eye lens dosimetry has been developed and this will improve the measurement and standardisation of the eye lens dose measurements for many years to come.
Development of type-tests and calibration procedures for personal dosimeters to be used in interventional radiology
A large number of international type-test and calibration standards are available for individual monitoring purposes. However, in the field of medical applications, there are several limitations that can be overcome by following some of the recommendations proposed in the framework of ORAMED.
A proposal for the type test and calibration of eye lens passive dosimeters especially used in IR/IC has been prepared. The proposal starts from the only one existing standard dealing with eye lens dosimetry using TLDs (ISO 12794) and proposes technical requirements for parameters such as, detection threshold, energy and angle responses. The recommended requirements are based on ISO 12794 and IEC 62387 and take into account the particular use of those dosimeters at IR/IC workplaces.
The proposal adopts the methodology developed in ORAMED project to reproduce and calculate in laboratory the operational quantity Hp(3). Conversion coefficients from air kerma to dose equivalent at 3 mm depth for RQR and ISO radiation qualities are presented in the proposal and the new water cylindrical phantom introduced by ORAMED is recommended for type-test and calibration. These proposals and coefficients should be taken into account by the international standard organisations when revising the relevant standards to allow a better estimate of Hlens.
The IEC 61526 standard is the international standard to be applied for type-testing of APDs. However, as pulsed radiation fields are not taken into account in existing standards, some information on the APD characteristics in pulsed field similar to those met at workplace are needed prior to using them in these fields.
The ORAMED project provides several guidelines to help the end-user in the selection and verification of the APD. Different sources of information are recommended: the results of the tests performed within the ORAMED project, specific ad-hoc tests performed by the manufacturers or some own tests performed in situ by the user.
The abovementioned proposal provides guidelines to calibration labs to perform calibration of eye-lens dosimeters and to derive the operational quantity from the reference air kerma value. Relevant recommendations should be used by medical physicists and radiation protection officers in order to select the best active personal dosimeter for their service and to make sure that their device will be appropriate for their use. The recommendations were first presented in the ORAMED workshop in Barcelona in January 2011 and are now available at the project website, as leaflets that can be distributed to target groups. Several manufacturers, present at ORAMED 2011, showed high interest on the results on APDs, both in laboratories and hospitals and encouraged the idea about having regular independent tests, as the ones performed in ORAMED.
Potential impact on regulation in radiation protection and more specifically in individual monitoring
The radiation protection recommendations and calibration guidelines are of interest to regulators and policy makers, on the one hand to disseminate good practices and on the other to up-date knowledge in radiation protection of medical staff and thus define reference levels and regulations for extremity dosimetry. Extremity dosimetry is far less developed in Europe than whole body dosimetry and there is a lack of systematic data. The situation is even worst in the case of eye-lens doses which are hardly ever measured or estimated. The results of the study highlight the importance of monitoring the wrists or fingers and eye lens.
In recent years there is growing epidemiological evidence of excess risk at the radiation doses measured within ORAMED, however, very often, dose assessment is not comprehensive in those studies. Moreover, the system of radiological protection is mainly based on excess risk of cancer induced by ionising radiation and only little information is available on non-cancer effects, such as cardio-vascular disease and lens opacities. The dose data base obtained in the ORAMED project will be of great use to contribute to new research on risks related to occupational doses. Representatives of most European radiation protection regulator bodies were present at ORAMED 2011.
Improvements in training on radiation protection for medical staff
Training material within the topics of the project has been prepared. Several proposals are available, specifically designed for different target-groups.
The material for medical staff consists of two modules (namely IR and NM) based on the experience of the project and taking advantage of images of practical situations as well as adapted to be used with interactive systems. Since there is already general training material available for trainers, we have prepared some recommendations on what training material could be used in each field and how this general information can be completed with the specific modules developed within ORAMED. IRSN has prepared some guidelines mainly based on IAEA modules available from the web site at http://rpop.iaea.org/RPOP/RPoP/Content/AdditionalResources/Training/1_TrainingMaterial/index.htm Material for dosimetry services and metrology labs was also prepared, by CEA, for dosimetry services and metrology labs, focussed on recommendations for the calibration of dosimeters to be used in interventional radiology.
SMU prepared a video for interventional radiology, showing how ORAMED measurements were performed. BfS prepared a video on Y-90 DOTA therapy. It includes good recommendations on radiation protection measures. Training material and videos are available at the ORAMED website and were presented during the ORAMED 2011 workshop.
Dissemination activities
The dissemination activities included the international workshop on 'Optimisation of radiation protection of medical staff', ORAMED 2011, which was organised from 20 to 22 January 2011, in the School of Industrial Engineering of Barcelona at UPC (Spain). Moreover, the project team participated in 10 scientific international conferences. In addition, presentations have been given to most national conferences in the topics studied in ORAMED and the final results were planned to continue to be presented during 2011 and beginning of 2012 in international conferences. Seven stages have been organised between the different groups, to let the young researchers learn from the experience of other partners and to harmonise work. ORAMED members also participated to five training activities organised by international institutions and professional societies. Finally, the ORAMED website was launched.
Project website: http://www.oramed-fp7.eu