If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
The efficiency of novel equipment to protect the staff in interventional cardiology, and particularly the eye lens and the brain, was investigated.
•
The dose reduction effectiveness of the caps, masks and drapes strongly depends on the design, exposure conditions and staff position.
•
Independent testing of radioprotective devices in realistic conditions of use is recommended.
Abstract
Purpose
To evaluate the effectiveness of currently available radioprotective (RP) devices in reducing the dose to interventional cardiology staff, especially to the eye lens and brain.
Methods
The performances of five RP devices (masks, caps, patient drapes, staff lead and lead-free aprons and Zero-Gravity (ZG) suspended radiation protection system) were assessed by means of Monte Carlo (MC) simulations. A geometry representative of an interventional cardiology setup was modelled and several configurations, including beam projections and staff distance from the source, were investigated. In addition, measurements on phantoms were performed for masks and drapes.
Results
An average dose reduction of 65% and 25% to the eyes and the brain respectively was obtained for the masks by MC simulations but a strong influence of the design was observed. The cap effectiveness for the brain ranges on average between 13% and 37%. Nevertheless, it was shown that only some upper parts of the brain were protected. There was no significant difference between the effectiveness of lead and lead-free aprons. Of all the devices, the ZG system offered the highest protection to the brain and eye lens and a protection level comparable to the apron for the organs normally covered.
Conclusion
All investigated devices showed potential for dose reduction to specific organs. However, for masks, caps and drapes, it strongly depends on the design, exposure conditions and staff position. Therefore, for a clinical use, it is recommended to evaluate their effectiveness in the planned conditions of use.
Fluoroscopically-guided interventional procedures associate a radiological imaging technique (using low energy X-rays) and a minimally invasive procedure with diagnostic or therapeutic aim. Using imaging enables to guide the surgeon and to do less invasive surgeries. Thus, interventional procedures in cardiology are techniques that have had widely spread in the last decades. Although interventional cardiology (IC) enables less invasive surgeries, there are concerns about the exposure of both the patient and the staff. Regarding medical staff working in IC, they are mainly exposed to scattered ionizing radiation.
The interventional staff receives the highest eye lens doses of ionising radiations in the medical sector, with cumulative eye lens doses ranging from a few mSv up to a few Sv over their entire career [
]. Since the early 2010s, several studies have reported an elevated incidence of eye lens opacities typically associated with ionising radiations among the interventional staff especially in cardiology [
], a disproportionate incidence of left-sided brain tumours (85 %) was observed. Given that the left side of the interventional cardiologist head is known to be more exposed to radiation than the right one, these findings raise concern.
Thus, there has been an increasing interest in radiation protection devices that offer shielding in particular to the head. Many devices are available to protect the staff: from frequently used equipment, such as lead glasses or ceiling-suspended screen, to drapes positioned on the patient or the ceiling- suspended cabins (e.g. Zero-Gravity (ZG) system). However, estimating the actual dose reduction effectiveness of those devices remains challenging because it can be strongly affected by their design and by the exposure conditions. Among other activities, the MEDIRAD project (Implications of medical low dose radiation exposure) aims to bridge gaps in the knowledge of staff radiation protection. In this framework, the effectiveness of currently available radioprotective (RP) devices in reducing the dose to interventional cardiology staff, especially to the eye lens and brain, was investigated. From an extensive literature review carried out at the beginning of the project, ten different RP devices were identified [
Dabin J, Domienik-Andrzejewska J, Huet C, Mirowski M, Vanhavere F. Report on effectiveness of protective devices for staff in interventional procedures. Deliverable 2.19, MEDIRAD project 2021. Available at : Documents download module (europa.eu).
]. A need for additional investigations was identified for five of them: masks, caps, patient drapes, staff lead and lead-free aprons and the Zero-Gravity suspended radiation protection system. As a matter of fact, for some of the RP devices, a limited number of publications was available: for the mask, only one publication presenting attenuation measurements in laboratory conditions [
] and for the lead-free apron no publication about effectiveness in clinical settings whether measured or simulated with MC. For others (drape and ZG system), the effectiveness was only investigated by means of clinical measurements [
Efficacy of MAVIG X-Ray Protective Drapes in Reducing Operator Radiation Dose in the Cardiac Catheterization Laboratory: A Randomized Controlled Trial.
In the framework of the MEDIRAD project, the performances of the five devices were assessed by means of Monte Carlo (MC) simulations and measurements on staff and phantoms. This article presents the results of the MC simulations for the five devices and of phantom measurements for two devices (mask and drape); results of the measurement campaigns on staff are presented elsewhere [
Dabin J, Domienik-Andrzejewska J, Huet C, Mirowski M, Vanhavere F. Report on effectiveness of protective devices for staff in interventional procedures. Deliverable 2.19, MEDIRAD project 2021. Available at : Documents download module (europa.eu).
Efficacy of MAVIG X-Ray Protective Drapes in Reducing Operator Radiation Dose in the Cardiac Catheterization Laboratory: A Randomized Controlled Trial.
]. The purpose of the present study was: (1) to assess the dose reduction effectiveness of RP tools, especially for the eye lens and the brain, (2) to identify the parameters influencing the dose reduction effectiveness (irradiation parameters as well as RP tool intrinsic parameters) and (3) to compare the”real” dose reduction effectiveness of the RP device (assessed at the level of the organs) to the one commonly estimated from measurements on staff.
Material and methods
Monte Carlo simulations were performed for the five RP devices to calculate their potential for reducing exposure of various organs at the settings commonly used during IC procedures. In addition, the correlation between the “real” effectiveness (calculated at the level of the organs) and the effectiveness as commonly estimated from measurements on staff (calculated at the level of the dosimeters) was investigated.
Interventional cardiology setup
A geometry representative of an interventional cardiology setup was simulated using MCNPX [
] MC codes. The examination settings and their values are listed in Table 1. Five X-ray beam projections, commonly used in clinical practice during IC procedures, were considered, namely postero-anterior (PA), left anterior oblique at 45° and 90° (LAO45 and LLAT) and right anterior oblique at 45° and 90° (RAO45 and RLAT). Two standing positions of the operator were considered: 40 cm and 70 cm from the centre of X-ray field on the right side of the patient, representing radial and femoral access, respectively. For the ZG device, only the position at 70 cm was simulated since it was not possible to model the position at 40 cm due to its bulkiness. Finally, two orientations of the operator’s head were modelled: forward and 30° away from the X-ray tube.
Table 1Examination settings and values used for the simulations.
Examination settings
Values
Use of RP device
With - without
Beam projections
PA, LAO45, LAO90, RAO45, RAO90
Operatoŕs distance from the X-ray field
40 and 70 cm
Operatoŕs head orientation
forward (0°) and 30° away from the X-ray tube
Field size at the detector (cm)
30 × 30: PA, LAO45 and RAO45 20 × 20: LLAT and RLAT
An X-ray spectrum corresponding to a tube voltage of 80 kV with a filtration of 3 mm Al was used. The X-ray photons were emitted from a point source into a cone and collimated into a square field of 20 × 20 cm2 or a 30 × 30 cm2, depending on the projection, at the entrance of the image detector. The source-to-image-detector distance was fixed (90 cm); the source-to-patient skin entrance distance and the patient-to-image-detector distance was set to 60 cm and 10 cm in PA, respectively, and could change, due to beam rotation, for the remaining projections. The image detector was modelled by a hollow lead box (2 mm Pb on the side, 5 mm Pb on top and containing air inside) with a 1.5 mm thick Al window on the entrance side above the patient. No RP room equipment was modelled since we wanted to investigate the effectiveness of the devices separately.
Numerical phantoms
Three kinds of numerical phantoms are available for MC simulations: stylized, voxel and hybrid phantoms. In stylized phantoms the organs are defined from a combination of mathematical equations which describe elliptical surfaces, cylindrical surfaces, etc. The advantages of these phantoms are smooth surfaces and flexibility – they can be modified by appropriate changes in parameters of the equations they are described with. However, the anatomical realism is limited, especially regarding the shape and the location of the organs. Voxel phantoms are created from CT or MRI of patients or corpses. Thus, these phantoms are representative of the anatomy, but they are hardly flexible and so can hardly be modified in terms of posture and morphology. Finally, hybrid phantoms are based upon non-uniform rational B-spline (NURBS) and/or polygon mesh surfaces. They combine the advantages of both stylized and voxel phantoms, namely flexibility and anatomical realism.
All in all, five different numerical phantoms were constructed from the three kinds of phantoms described above and used in the simulations: one for the patient and four for the interventional cardiologist, selected on a case-by-case basis according to the type of protection being studied. A single simple stylized phantom was used for the patient [
]. There is no distinction between the different organs, only a simplified skeleton and the lungs are modelled in the patient. Regarding the cardiologist, one phantom was used for the study of masks and caps, a second for the study RP drapes; a third for the study of aprons; and a fourth for the study of the ZG device. These phantoms are detailed in Table 2. For the study of masks and caps, the phantom constructed by Silva et al. [
] was used. In addition, for this study, the voxelised head has been specially modified to include the detailed ICRP eye lens model, which considers the sensitive and insensitive volumes of the eye lens separately [
] (Fig. 1). Regarding ZG device, a voxelised phantom was created from the realistic anthropomorphic flexible polygonal mesh phantom developed by Lombardo et al. [
Kramer R, Zankl M, Williams G, Drexler G. The calculation of dose from external photon exposures using reference human phantoms and Monte-Carlo methods. Part I. The male (ADAM) and female (EVA) adult mathematical phantoms GSF Bericht S-885 1982.
Several masks are commercially available. Two of them (VIS400 face mask (Longkou Sanyi Medical Device Co., China) and Full face style mask (Philips Safety products, USA)) were chosen (Fig. 2) (named M1 and M2). They are both composed of an acrylic and lead mixture whose exact proportions are unknown. As such, they were modelled using the lead equivalence of 0.1 mm announced by the manufacturer. However, their shapes are quite different which could influence their protective capabilities for organs located in the head. Thus, a third mask (named M1L), not available, was also modelled as M1 with a 7.8 cm longer screen, so that it had the same length as M2 in order to study the effect of the length.
Fig. 2Pictures of the commercial masks modelled in the MC simulations a: M1, b: M2, c: frontal view of the cardiologist model with the M1 mask, d frontal and e: lateral view of the cardiologist model without the mask; dosimeters are represented in red (soft tissue dosimeters positioned above and under the mask) and blue (eye lens). Dosimeters were enlarged for visualisation purpose.
For M1 and M2, four pairs of dosimeters were placed above and below the mask at four different positions (left lower and higher rim and right lower and higher rim) in order to evaluate its effectiveness as commonly estimated from measurements on staff. Dosimeters were modelled by cylinders (0.16 cm3) made of soft tissue material. Finally, an eye lens (Hp(3)) dosimeter was modelled on the left temple of the numerical phantom (Fig. 2).
Caps
The shape of the commercially available caps is quite similar from one model to another. In order to study the effect of the cap composition on its effectiveness, one cap model with two different compositions (pure lead and a lead-free metal alloy) was modelled. The cap model is based on a commercial surgical model (RADPAD No Brainer X-ray Protective Surgical Cap; Worldwide Innovations & Technologies) which is 12 cm high and covers the head obliquely from above the eyebrows and the nape (Fig. 3). The top of the head is unshielded.
Fig. 3a: Picture of the cap modelled (RADPAD No Brainer X-ray Protective Surgical Cap; Worldwide Innovations & Technologies; source: http://www.varaylaborix.com), b: frontal view of the cardiologist model with the cap, c: frontal and d: lateral view of the cardiologist model without the cap; dosimeters are represented in red (soft tissue) and blue (eye lens). Dosimeters were enlarged for visualisation purpose.
Randomized Controlled Trial of Radiation Protection With a Patient Lead Shield and a Novel, Nonlead Surgical Cap for Operators Performing Coronary Angiography or Intervention.
]. Three pairs of dosimeters made of soft tissue material were simulated above and below the cap. They were placed near the left temple, at the forehead between the eyes and at the end of the left eyebrow (Fig. 3). Finally, an eye lens (Hp(3)) dosimeter was modelled directly on the left temple of the numerical phantom.
Patient drapes
A commercially available lead-free drape (0.375 mm Pb eq) was modelled. The material composition was obtained from the manufacturer under a non-disclosure agreement. Although the manufacturer advised different drape positions depending on the access route (represented in the MC simulations by the interventional cardiologist position with respect to the X-ray beam centre), this could not be implemented. Positioning the drape closer to the patient chest as advised for a radial or brachial access route would have resulted in the drape interacting with the primary beam. Only one drape position was therefore considered for the simulation set-ups (Fig. 4). An eye lens (Hp(3)) dosimeter was modelled on the left temple and a whole-body (WB, Hp(10)) dosimeter on the left side of the chest above the apron, in order to evaluate the drape effectiveness as derived from measurements on staff in clinical studies. In addition, to assess the effect on the skin extremities, thin tallying regions (1-mm thickness) were defined at the palm and forearm surface facing the patient.
Fig. 4Top view from the ORNL-ORAMED phantoms used for investigation of drape effectiveness. The cardiologist is positioned at a: 40 cm and b: 70 cm from the X-ray beam centre. The radioprotective drape is coloured in black; the image detector in purple.
Three full-body apron models with different compositions were studied. Pure lead (LA) and two lead-free compositions, based on manufacturer data (LFA1) and on data from experimental characterization (LFA2) [
Aral N, Duch MA, Ardanuy M. Material characterization and Monte Carlo simulation of lead and non-lead X-Ray shielding materials. Radiation Physics and Chemistry 2020; 174. doi:10.1016/j.radphyschem. 2020. doi:10.1016/j.radphyschem.2020.108892.
], were modelled (Table 3). All aprons had the same design, only the effect of the material composition and/or the actual thickness were considered. For each apron, a thyroid collar with the same thickness and composition was also modelled. Finally, a pair of dosimeters was placed on the left side of the chest above and below the apron.
Table 3Characteristics of the lead and lead-free aprons investigated.
Zero-Gravity suspended radiation protection system
The ZG (Biotronik, Germany) was modelled based on the information provided by the manufacturer: an apron with lead thickness varying between 0.5 and 1 mm for the apron and the arm flaps, depending on the location, and 0.5 mm for the face shield. The model was then included in a software enabling fast generation of MCNP input files (Fig. 5) [
]. As already mentioned, due to its bulkiness it was not possible to simulate the operator at 40 cm from the source. A WB (Hp(10)) dosimeter was added on the left side of the cardiologist chest, above the apron when used, and below the ZG otherwise.
Fig. 5Flexible numerical phantom equipped with a: apron and b: the ZG suspended system (screenshot from the software tool used for MCNP input file generation). The yellow halo on the right arm, the blue sphere on the left one and the green cylinder on the left shoulder and the cross on the ground are elements used for visualization and modification of the phantom posture. They are not modelled in the simulations.
Monte Carlo simulations settings and data analysis
The default MCNP physical parameters for photon transport (i.e. including fluorescence and energy cut-off set to 1 keV) and ENDF/B-VI libraries were used. The DXTRAN variance reduction method was used [
]. This method enables to increase the number of particles in the region of interest. In this study, the DXTRAN sphere was placed around the operator’s head.
The energy deposition, in MeV/g, in different organs or tissues and dosimeters was calculated by using MCNPX/MCNP F6:p tally. This tally uses the kerma approximation (energy from electrons created by photon interactions is assumed to be deposited locally). Some simulations were performed with and without this approximation (tally *F8) in order to verify that it was valid in our case (thin volumes for dosimeters).
For the drape, the apron and the ZG system, the effective dose was calculated applying ICRP 103 [
ICRP. The 2007 Recommendations of the International Commission on Radiological Protection. ICRP publication 103. Ann ICRP. 2007; 37(2-4):1-332. doi: 10.1016/j.icrp.2007.10.003.
] formalism on the organ and tissue doses. The effective dose was not considered for the cap and the mask as with these RP devices only the head region might be protected.
The dose reduction effectiveness (expressed in %), also called effectiveness in the paper, of the RP device for the different organs investigated and for a given projection was calculated as the difference between the absorbed dose without the device (the control dose, DC) and the dose with the RP device (DRP), normalized to the dose without the device:
(1)
Attenuation (expressed in %) was calculated for dosimeters placed above and under the caps and the masks as the difference between the dose absorbed above (DA) and below the RP device (DB), normalized to the dose above the device:
(2)
Phantom measurements
In order to further validate the results of the MC simulations, results were compared with measurements on anthropomorphic phantoms in clinical configurations. Results of such experiments being already available in the literature for the caps [
Grabowicz W, Masiarek K, Tomasz Górnik T, Grycewicz T, Brodecki M, Dabin J et al.. The effect of lead free cap on the doses to the head of interventional cardiologists working in haemodynamic room, submitted.
] and a PMMA phantom were set to represent the cardiologist and the patient, respectively. The RA phantom represents a 1.75 m tall and 73.5 kg male adult, without arms or legs, and is composed of multiple slices of tissue-equivalent material. The RA phantom was placed next to the operating table in the location corresponding to the position of cardiologist performing a real procedure. The head of the RA phantom was filled with 138 TLDs. This enabled the estimation of the dose to the brain (including cerebellum) and to the complete head region. For the evaluation of the exposure at the level of the left temple and both eye lens, three loose dosimeters were placed at each location during single exposure.
Measurements were performed on a C-arm Philips BV unit (Philips, The Netherlands). For each configuration, measurements were done sequentially with and without the RP device. Measurements were performed using three (PA, LAO 30 and LLAT) to four (PA, LAO 30, LLAT and RAO 30) beam projections for the drape and the mask respectively. Those beam projections were selected because of their significant frequency of use (PA, LAO 30 and RAO 30) and their potential for high staff exposure (LLAT). When the beam projection was changed, all other geometrical parameters were kept constant. Additional filtration and tube voltage of the X-ray unit were selected in the former case automatically while in the latter manually; the corresponding values were 3.0 mmAl and 90 kV. Although it was strived to achieve identical tube output levels (PKA) for similar configurations, measurement results were normalised to the PKA. Effectiveness of the RP device was calculated as per Equation (1).
A picture of the set-up is presented in Fig. 6. The distance of the RA phantom from the centre of the beam incident on the patient was 60 cm and the field size was 20 cm at the image detector position.
Fig. 6Measurement of the effectiveness of a mask (model M1) and a lead drape (LITE TECH, Inc., USA) on a RA phantom in a clinical set-up: a: position of dosimeters on the left temple and the eye lens, b: LAO 30 projection with neither mask nor drapes, c:LAO 30 projection with the mask d: LAO 30 projection with the lead drape.
For masks and caps simulations, statistical uncertainties were below 1 % for the brain, between 1 % and 3 % for eye lens and eye lens (Hp(3)) dosimeter and below 5 % for the other dosimeters. For the aprons, statistical uncertainty on effective dose was less than 1 %. For the drape simulations, uncertainties were below 1 % for the brain; below 3 % and 8 % for the eye lens and eye lens dosimeter, respectively; and below 4 % and 1 % for the WB dosimeter and the effective dose, respectively. For the ZG simulations, the uncertainties were less than 2 % for the brain; for the eye lens and the WB dosimeter, the uncertainties were below 1 % when the ZG was not used and up to 7 % and 12 % when the ZG was used, respectively; and below 1 % for the effective dose.
Masks
M1 and M2 masks attenuation obtained from the dosimeters placed at different locations above and below the mask averaged over the five simulated projections is given in Table 4. While attenuation for M1 is similar whatever the location of the dosimeters, a lower attenuation is observed for M2 for the dosimeters located on the right higher rim, in particular at 70 cm. However, the attenuation for a given distance is comparable between the two masks.
Table 4M1 and M2 masks attenuation obtained from the dosimeters placed at different locations above and below the mask averaged over the five simulated projections at 40 cm and 70 cm.
M1 mask effectiveness in reducing the doses to the sensitive volume of the eye lens, brain and Hp(3) dosimeter obtained from MC calculations for the different projections and the two distances investigated (head at 0°) is presented in Fig. 7. The effectiveness from phantoms measurements at 60 cm is presented in Table 5. For all projections, a better simulated effectiveness is obtained at 70 cm compared to 40 cm. Additional simulations performed at 50 cm and 60 cm for PA projection indicate that the effectiveness in reducing the doses to Hp(3) dosimeter increases from 3 % to 19 % between 50 cm and 60 cm and from 19 % to 59 % between 60 cm and 70 cm. Regarding the whole brain, it increases by step of 10 % between 40 cm and 70 cm, ranging from 11 % to 39 %.
Fig. 7M1 simulated dose reduction effectiveness for brain, eye lens and Hp(3) dosimeter for the different projections and the two distances investigated (head at 0°). The brain includes the white matter and the lobes.
Whatever the projection and the distance, M1 effectiveness in protecting the eye lens is poor. In the case of the right eye lens, the effectiveness obtained from phantom measurements is negative for LLAT and this is rather due to statistical reasons than the actual dose increase due to the presence of the mask. At 70 cm (Fig. 7), the effectiveness in reducing the doses to Hp(3) dosimeter is around 65 % which is more than 8-fold the reduction for the sensitive volume of the eye lens, and at 60 cm (Table 5) it is on average 7-fold. Effectiveness for the brain averaged over the projections is around 12 % at 40 cm and 60 cm and 43 % at 70 cm.
M1, M1L and M2 simulated effectiveness for eye lens, brain and Hp(3) dosimeter for the head at 0° and 30° and the two distances investigated is presented in Fig. 8. Effectiveness is averaged over the five projections. Regarding M1, no significant difference is observed between the head at 0° and the head at 30°, except for the shielding effectiveness to the brain which is enhanced by 15 % at 40 cm when the head is rotated. M1L, which is a lengthened version of M1, is much more efficient than M1 for the brain (62 %), the eye lens (64 % (left)) and Hp(3) dosimeter (84 %) at 40°cm as well as at 70 cm, especially for the brain and the eye lens. Finally, with the head at 0°, M2 is very efficient for eye lens (70 % (left)) and Hp(3) (60 %) and to a lesser extent for the brain (20 %) at 40 cm and 70 cm. Besides, it drops down to a few percent with the head rotated at 30° for both distances.
Fig. 8Simulated dose reduction effectiveness for M1, M1L and M2 at the level of the brain, eye lens and Hp(3) dosimeter, considering two operator head positions (0 and 30°) and distances (40 and 70 cm).
Simulated dose reduction effectiveness averaged over the five projections for a same operator distance and head position with the lead and lead-free caps is given in Table 6; little difference was observed between the projections (standard deviation always < 5 %, not shown). Results obtained for the two caps are very similar. The effectiveness for the brain is low, especially at 40 cm. It is improved when the head is rotated at 30°. However, whatever the head orientation, some parts of the brain are more protected than others (Fig. 9). As expected, both caps are ineffective to reduce the eye lens dose (less than 1 %).
Table 6Simulated dose reduction effectiveness obtained for the lead and lead-free caps averaged over the five projections at 40 cm and 70 cm.
Depending on the dosimeter location, the calculated cap attenuation differs considerably (Fig. 10). The attenuation obtained from the dosimeters located near the left eyebrow and on the left temple is quite similar (around 88 %) and does not vary much whatever the projection, the head position (0° or 30°) and the operator position (40 cm or 70 cm). By contrast, the attenuation obtained from the dosimeters located between the eyes is less than the one obtained from the two other dosimeter positions and varies considerably according to the projection, the head orientation and the operator location. A lower attenuation is obtained with LLAT projection for the head at 0° for both operator positions. Finally, with the head rotated at 30° the cap attenuation drops, especially for LAO and LLAT (values below 0) for both operator positions.
Fig. 10Lead cap attenuation obtained from the dosimeters placed above and below the cap, at 40 cm and 70 cm and for the head at 0° and 30°.
Simulated dose reduction effectiveness obtained for the lead (LA) and the two lead-free aprons (LF1 and LF2) for the different projections at 40 cm and 70 cm is given in Table 7. The lead-free aprons show no noticeable difference compared to lead aprons, irrespective of the simulation conditions. For the three aprons, the larger the distance the lower the effectiveness. These slight decreases in effectiveness are mostly due to decreased effectiveness for some organs like lungs, oesophagus, brain and heart (Table 8). The dose reduction obtained for the dosimeter placed below the apron on the phantom torso is close to the effective dose reduction. The mean relative difference is 3.5 % for the lead apron and 6 % for the lead-free aprons. However, in some cases, this difference changes drastically. For instance, for LLAT at 70 cm, the dose reduction obtained for the dosimeter is around 33 % for the three aprons whereas it is around 79 % for the effective dose.
Table 7Simulated dose reduction effectiveness obtained for the lead and lead-free aprons for the different projections at 40 cm and 70 cm.
In a fixed clinical setting, the dose reduction effectiveness for the head region, resulting from MC calculations and phantom measurements, was low. Different projections resulted in simulated effectiveness around 3 % on average for the brain and the eye lens whereas measured effectiveness was below 7 % on average for the eye lenses and not meaningful at the level of the brain. The simulated effectiveness was slightly higher at 70 cm (2.9 %, 3.8 % and 4.4 % for the brain, the left and the right eye, respectively) compared to 40 cm (1.6 %, 2.3 % and 2.5 % for the brain, the left and the right eye, respectively). The drape was more efficient for RAO45 projections for both operator positions: for that projection, the effectiveness ranged from 5.9 % to 10.5 %. No statistical difference was observed between effectiveness for the eye lens (Hp(3)) dosimeters and for the actual eye lens. Specific brain regions show dose reduction trends, according to the combination of operator distances and projections, which are comparable to the trends observed for the complete brain.
Significant dose reduction was observed for organs situated in the direct vicinity of the drape, such as the hand skin and, to a lesser extent, the forearm skin (Table 9). In particular, an effectiveness of 62 % and 30 % is observed to the left and right hands, respectively, when they are directly situated above the drape (40 cm). A decrease up to 72 % and 40 % is observed for the left hand in RAO45 and for the right hand in RLAT, respectively. In contrast, effectiveness for the left and right hands is only 18 % and 8 % when the hands are further away from the drape (70 cm). Reduction to other extremities, not in the close vicinity of the drape, is much more limited (forearm) or not observed (leg).
Table 9Simulated dose reduction effectiveness of the drape for the arm skin and the hand skin at 40 cm and 70 cm.
The drape also appears to decrease the dose to some organs situated in the abdomen region, although already protected by the apron, but only for right RLAT projections. For instance, an effectiveness of 18 % and 48 % is observed for the uterus in RLAT at 40 cm and 70 cm, respectively. Absolute dose reduction, however, is small since the lead apron offers an important effectiveness as evidenced by the results in Table 8.
Zero Gravity
Results of Monte Carlo simulations for the ZG are presented in Fig. 11 and Table 10. Different projections resulted in average dose reduction of at least 90 % for all the organs in the head and neck region (Fig. 11) while for organs normally covered by the apron (Table 10) the average dose reduction ranged from 23 % to 81 %. Negative dose reduction (dose increase) was observed for the left lung and the stomach.
Fig. 11Simulated dose reduction effectiveness of ZG to various organs in the head region for various projections at 70 cm from the primary beam.
Mask attenuation assessed by means of MC simulations obtained from dosimeters placed above and below the mask was very high (82 %) and similar whatever the projections. This attenuation is however not representative of the dose reduction (effectiveness) for the organs and tissues located in the head. Indeed, nearly all the radiation which directed to the dosimeters placed below the mask reaches the mask and are thus attenuated. However, the path of radiation reaching the organs located in the head does not necessarily cross the mask. The magnitude of this effect depends on the mask design and length. For instance, M1 dose reduction effectiveness for the eye lens is very low, either by Monte Carlo simulation (less than 8 %) or by phantom measurements (below 10 %). This suggests that most of the radiations reaching the eye when wearing M1 do not pass through the mask but through the gaps between the mask and the face and thus are not attenuated as shown by Koukorava et al. [
] for RP glasses. A better protection of the eye lens is obtained with M1L which is a version of M1 extended towards the chin. This is also confirmed, on one hand, by enhanced eye lens effectiveness obtained with M2 which is also long, and on the other hand, by the higher eye lens effectiveness obtained with M1 at 70 cm compared to 40 cm. At larger distances, a smaller proportion of radiation comes from underneath the mask. These results are in agreement with those of Koenig et al. [
] who measured on phantoms a dose reduction to the eye lens of 70 % for a mask similar to M2 while a non-significant dose reduction of 20 % was obtained for a mask similar to M1. Regarding dose reduction to the brain with the head at 0°, it was found to be better with M1L at the two distances, followed by M2 and M1 at 40 cm; the contrary was found at 70 cm. Once again, these differences can be explained by the length and the shape of the masks. Longer masks are more efficient at shorter distances. However, as the distance from the source increases, effectiveness of M2 decreases due to its particular shape on the sides. This is even more noticeable when the head is rotated at 30° where its effectiveness drops down to a few percent.
From MC simulations, also the dose reduction at the Hp(3) dosimeter is not representative of reduction to the eye lens, in particular for the short mask M1 at 70 cm. Relying on effectiveness values determined using Hp(3) dosimeters could thus lead to a severe underestimation of the real eye lens dose. Similar findings are obtained with phantom measurements: Hp(3) effectiveness ranged from 33 % to 50 % while eye lens effectiveness was less than 10 %. The location of Hp(3) dosimeter can explain this difference: it was located on the left temple closer to the mask surface than the left eye. The additional calculations performed with the operator at 50 cm and 60 cm with M1 indicated that the effectiveness at the Hp(3) dosimeter increased between 60 cm and 70 cm. This suggests that between these two distances, the majority of the radiation which reached the dosimeter from underneath the mask came through the mask.
Finally, comparable efficiencies (around 40 %) are obtained with M1 mask for the whole brain both with simulations and phantoms measurements.
Caps
From MC simulations, an average cap attenuation of 86 % was obtained from the dosimeters positioned directly above and below the lead and lead-free caps. This is in agreement with the 85 % obtained by Uthoff et al. [
] from measurements on staff and with the attenuation of 81 % and 86 % obtained from phantom measurements at the left eyebrow and left temple respectively [
Grabowicz W, Masiarek K, Tomasz Górnik T, Grycewicz T, Brodecki M, Dabin J et al.. The effect of lead free cap on the doses to the head of interventional cardiologists working in haemodynamic room, submitted.
]. The attenuation level, however, could be lower (around 50 % on average) in clinical conditions from dosimeters worn above and below a lead-free cap by medical staff [
Grabowicz W, Masiarek K, Tomasz Górnik T, Grycewicz T, Brodecki M, Dabin J et al.. The effect of lead free cap on the doses to the head of interventional cardiologists working in haemodynamic room, submitted.
]. The cap effectiveness obtained from the dosimeters located on the middle of the forehead was less than the one obtained from the two other locations (left eyebrow and left temple). For the middle of the forehead, the effectiveness obtained for LLAT projection by simulations with the head rotated at 30° was below 0. That means that the dose received by the dosimeter below the cap exceeded the one received by the dosimeter above the cap. A similar behaviour was noticed by Grabowicz et al. [
Grabowicz W, Masiarek K, Tomasz Górnik T, Grycewicz T, Brodecki M, Dabin J et al.. The effect of lead free cap on the doses to the head of interventional cardiologists working in haemodynamic room, submitted.
]. This suggests that the radiation reaching the dosimeter may come from underneath the cap or may pass through the dosimeter located inside the cap before passing through the dosimeter outside. For the first hypothesis, it could be due to the fact that the cap does not fit perfectly the head of the numerical phantoms while for the second one it suggests that radiations come from the left side or the back of the head.
In our simulation study, lead and lead-free caps have a potential for significant dose reduction to the brain (up to more than 35 %). However, this strongly depends on the relative position of the physician with respect to the primary X-ray field: with the configuration with the head at 0° the average reduction was only about 13 % at 40 cm whereas it was 37 % at 70 cm. This is in agreement with Silva et al. [
] who calculated an average dose reduction to the brain between 6 % and 15 % at 40 cm. This is also the same magnitude as the 7 % effectiveness obtained by means of phantom measurements with TLD [
Grabowicz W, Masiarek K, Tomasz Górnik T, Grycewicz T, Brodecki M, Dabin J et al.. The effect of lead free cap on the doses to the head of interventional cardiologists working in haemodynamic room, submitted.
] thanks to measurements with radiochromic films in an anthropomorphic phantom. In addition, our MC study revealed that the larger the distance, the better the effectiveness. As suggested with the mask, a smaller proportion of radiations may come from underneath the cap at larger distances.
The MC simulations also showed that attenuation calculated from dosimeters located directly above and below the caps is a poor estimator of the brain protection. Indeed the attenuation derived from the dosimeters (more than 80 % on average) severely overestimated the dose reduction in most configurations (on average between 13 % and 37 % at 40 cm and 70 cm, respectively with the head at 0°). This is in agreement with Silva et al. [
] who reported that no dosimeters placed under a cap was appropriate to estimate the brain dose reduction because, on average, only 5 % of the brain exposure penetrated the head through the forehead when the physician was at 40 cm from the X-ray field.
Aprons
Comparable attenuations were obtained for lead and lead-free aprons by means of numerical simulations. This is not in agreement with the study of Schlattl et al. [
] who, from MC calculations of air kerma below an apron in a broad primary beam, reported that the shielding capability of lead-free materials composed of Sn and Sn/Bi is much lower than that of lead. This was attributed to the high proportion of low-energy photons created by fluorescence in tin. However, as several parameters are different between our study and the one of Schlattl et al. [
] (primary beam versus scattered beam, Sn/Bi versus Sb/Bi, kV and filtration, etc.), comparison of the findings is not straightforward. Effective dose reductions obtained from our MC simulations were also quite similar for lead and lead-free aprons. The reduction was around 3 % less with the lead-free aprons. Schlattl et al. [
] reported around 3 % effective dose increase with tin/bismuth shielding at 75 kV compared to lead shielding. This moderate increase, compared to the one observed for air kerma by the same authors, was attributed to the fact that low-energy photons created by fluorescence cannot penetrate deeply into the body. Therefore, they observed a distinctive dose increase only in organs located very superficially, as glandular breast tissue for instance.
The decrease of effectiveness when the distance from the source increases, noticed with numerical simulations, may be attributed to two factors. On one hand, moving away from the source, the shadow created by the apron on the head and neck region could be reduced. This would explain the lower effectiveness obtained for the brain and the thyroid at 70 cm. On the other hand, there are some holes in the apron for the arms. They are not wide but their height can reach 20 cm; thus when the distance from the source increases, the trunk and organs like the heart are more exposed because of the apron holes. It should also be kept in mind that the absolute organ dose usually decreases further away from the primary X-ray field. For instance, the dose to the organs included in the effective dose calculations decreases on average by about 60 % at 70 cm with respect to 40 cm.
Except for LLAT at 70 cm, there was a good agreement between the reduction for the dosimeters and the effective dose. Therefore, the dose reduction to the dosimeters is a good approximation of the apron effectiveness for the effective dose.
Patient drapes
From simulations, a lead-free or lead drape positioned on the patient appeared to be an efficient solution to reduce the dose to the left hand and, possibly, to the right hand. The dose reduction to locations such as the brain, the eye and the WB dosimeter, however, was very low or inexistent. Phantom measurements confirmed these results (from the present study and from [
]). However, considerable variation in the drape effectiveness at the level of the eye lens and the WB dosimeter was observed in clinical studies in IC. From measurements on staff, Musallam et al. [
A randomized study comparing the use of a pelvic lead shield during trans-radial interventions: Threefold decrease in radiation to the operator but double exposure to the patient.
Efficacy of MAVIG X-Ray Protective Drapes in Reducing Operator Radiation Dose in the Cardiac Catheterization Laboratory: A Randomized Controlled Trial.
], 50 % and 53 % decrease in left eye dose and WB dose was measured on average, respectively. The considerable difference in dose reduction effectiveness between the clinical measurements and the MC simulations has to be investigated further.
Zero Gravity
As demonstrated by MC simulations the ZG is equivalent to the lead and lead-free aprons for the organs and dosimeters usually covered by the latter. Similar results were obtained from phantom and staff measurements [
], MC simulations showed an important effectiveness for the sensitive organs in the head region. In contrast with the mask and the cap which partially cover the head, the ZG effectiveness was rather homogenous for all tissue and organs in the brain region. Reduction to the eye lens dose ranged from 83 % to more than 95 % depending on the projection. This is in agreement with the study of Savage et al. [
] who reported 94 % reduction during various types of interventional procedures and more effective than the 50 % dose decrease to the eye lens reported by Haussen et al. [
A prospective case control comparison of the ZeroGravity system versus a standard lead apron as radiation protection strategy in neuroendovascular procedures.
] during interventional neuroprocedures. The MC simulated results are also consistent with the validation measurements performed on an anthropomorphic phantom showing dose reduction to the brain and eye lens from 66 % up to 96 % [
] who reported a dose reduction to the eye lens of 70 %.
In addition, the MC simulations showed that the ZG system reduced the dose to organs normally protected by the apron by 5 % to 98 %, probably because the apron of the ZG system has a higher lead thickness than a conventional apron. These additional reductions lead to decrease in effective dose of around 80 % for the different projections. One should keep in mind, however, that the absolute dose to those organs is usually low when a lead apron is used and can be 100 times lower than the dose to the unprotected left eye lens for instance.
Surprisingly, simulations also showed a dose increase for some internal organs situated on the left side of the interventional cardiologist body in RLAT projections. For instance, 18 % dose increase was observed at the lung and the stomach. Aside from the low magnitude of the dose to those organs covered by the lead apron as already mentioned, this was probably caused by an inaccurate modelling of the ZG, in particular the tissue providing shoulder and upper arm protection. While in reality this protection can easily fold to provide complete coverage of the upper arm during movements, this is not the case for the model used in the simulations, leaving a fixed air gap between the arm and the protection.
Recommendations
Formulating recommendations is not straightforward since the effectiveness of most devices strongly depends on their design and the exposure conditions. For those personal devices that are directly aimed at protecting the head region, such as the masks, an extended face coverage and the smallest possible gap between the face and the device are crucial parameters to ensure adequate protection. Few devices, such as the ZG suspended system, can offer significant dose reduction in most circumstances. From a practical point of view, other factors, such as ergonomics, ease of use and ease of asepsis should also be taken into consideration when choosing a RP device. The results of the present study were used for producing device-specific recommendation in the form of pros and cons [
Only the dose reduction effectiveness was investigated in this study. In addition to the effectiveness, the absolute dose should be considered when evaluating the performances of a RP device: low dose reduction effectiveness for a specific tissue which is exposed to a very low dose is not of concern for radiation protection, while low dose reduction to a highly exposed tissue is of concern.
Besides, not all combinations of factors that can adversely affect the effectiveness of the RP devices were investigated. Although various configurations were simulated using MC calculations and phantom measurements, only a limited number of RP device models and designs were tested; and only one X-ray beam quality and a limited number of geometry set-ups were considered for the simulations. A comparison of the effectiveness of mask and glasses to protect the eye lens would also be of interest but is out of scope of the present study. Moreover, the X-ray beam quality used in MC simulations and phantom measurements slightly differ in high voltage (80 kV vs 90 kV, respectively) which had an additional input to observed differences in the effectiveness of RP devices obtained with the use of both methods. Some elements (walls and C-arm for instance) were also not modelled in MC simulations as their contribution to scatter radiation is considered to be negligible. Therefore, the effectiveness of the protective devices deserves to be further investigated for other procedure types with different access routes, staff’s orientations and anatomies.
The lead equivalence of the RP device should not be overlooked either. The attenuation characteristics as stated on the equipment label are usually indicative of the lead equivalence of the material (or the resulting X-ray attenuation) in a direct X-ray beam with a specific energy spectrum but they are not likely representative of the dose reduction to any specific organ in clinical conditions [
]. This is of even greater concern for lead-free and light lead material because the material properties cannot be easily extrapolated to other exposure conditions without information on the material composition, which is usually not known by the user. Not all simulations could be performed with the elemental composition since this was not always communicated by the manufacturers. In such cases, the lead equivalence was used; however, it can be considerably lower than the one stated on the device label [
Finally, the effectiveness of the RP devices was simulated when used individually. However, this is seldom the case in practice. MC simulations could lead to a better understanding of the combined effects and explain discrepancies found in various studies between simulations and clinical measurements.
Conclusion
The dose reduction effectiveness of five RP devices to reduce staff dose were assessed using MC simulations. Phantom measurements were also performed for masks and drapes to complete the data available in the literature. Lead and lead-free caps and masks have a potential for protecting – some parts of - the brain, while masks could protect the eye lens as well. However, their effectiveness strongly depends on the exposure conditions and the device design, potentially leaving staff unprotected without realising it. Lead-free aprons can offer comparable protection to a lead apron and the ZG system offers a considerable dose reduction to all organs covered. Whatever the RP device, it is advisable to always (re)evaluate the effectiveness in the planned conditions of use.
Funding
The Medirad project has received funding from the Euratom research and training programme 2014–2018 under grant agreement No 755523.
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
References
Struelens L.
Dabin J.
Carinou E.
Askounis P.
Ciraj-Bjelac O.
Domienik-Andrzejewska J.
et al.
Radiation-Induced Lens Opacities among Interventional Cardiologists: Retrospective Assessment of Cumulative Eye Lens Doses.
Dabin J, Domienik-Andrzejewska J, Huet C, Mirowski M, Vanhavere F. Report on effectiveness of protective devices for staff in interventional procedures. Deliverable 2.19, MEDIRAD project 2021. Available at : Documents download module (europa.eu).
Efficacy of MAVIG X-Ray Protective Drapes in Reducing Operator Radiation Dose in the Cardiac Catheterization Laboratory: A Randomized Controlled Trial.
Kramer R, Zankl M, Williams G, Drexler G. The calculation of dose from external photon exposures using reference human phantoms and Monte-Carlo methods. Part I. The male (ADAM) and female (EVA) adult mathematical phantoms GSF Bericht S-885 1982.
Randomized Controlled Trial of Radiation Protection With a Patient Lead Shield and a Novel, Nonlead Surgical Cap for Operators Performing Coronary Angiography or Intervention.
Aral N, Duch MA, Ardanuy M. Material characterization and Monte Carlo simulation of lead and non-lead X-Ray shielding materials. Radiation Physics and Chemistry 2020; 174. doi:10.1016/j.radphyschem. 2020. doi:10.1016/j.radphyschem.2020.108892.
ICRP. The 2007 Recommendations of the International Commission on Radiological Protection. ICRP publication 103. Ann ICRP. 2007; 37(2-4):1-332. doi: 10.1016/j.icrp.2007.10.003.
Grabowicz W, Masiarek K, Tomasz Górnik T, Grycewicz T, Brodecki M, Dabin J et al.. The effect of lead free cap on the doses to the head of interventional cardiologists working in haemodynamic room, submitted.
A randomized study comparing the use of a pelvic lead shield during trans-radial interventions: Threefold decrease in radiation to the operator but double exposure to the patient.
A prospective case control comparison of the ZeroGravity system versus a standard lead apron as radiation protection strategy in neuroendovascular procedures.
00020-0&id=gr1.jpg){kind=link}
00020-0&id=gr2.jpg){kind=link}
00020-0&id=gr3.jpg){kind=link}
00020-0&id=gr4.jpg){kind=link}
00020-0&id=gr5.jpg){kind=link}
00020-0&id=gr6.jpg){kind=link}
00020-0&id=gr7.jpg){kind=link}
00020-0&id=gr8.jpg){kind=link}
00020-0&id=gr9.jpg){kind=link}
00020-0&id=gr10.jpg){kind=link}
00020-0&id=gr11.jpg){kind=link}