ICRP Publication 139: Occupational Radiological Protection in Interventional Procedures

(f) X-ray image-guided therapeutic interventions such as radioembolisation with ⁹⁰Y-labelled microspheres [selective internal radiation therapy (SIRT)] are an alternative method to treat patients with unresectable primary or secondary liver tumours. Several hospitals are exploring the use of real-time positron emission tomography (PET)-CT guidance during interventional procedures, such as for biopsies and/or radiofrequency ablations. ¹⁸F-FDG PET-CT imaging is performed within the suite to identify where an embolisation or biopsy should be performed, to check on effectiveness of interventions, and for early detection of residual disease (e.g. after radiofrequency ablation, so that ablation can be repeated, if necessary, in order to obtain the maximum therapeutic benefit).

(h) The equivalent dose to the lens of the eye has received increased attention as evidence has become available that cataract development may have a much lower threshold for occurrence than was historically believed. The Commission’s recommendations have lowered the equivalent dose limit for the lens of the eye from 150 mSv year⁻¹ to 20 mSv year⁻¹, averaged over defined periods of 5 years with no single year exceeding 50 mSv. The nature of interventions guided by radiological imaging is such that, without protective measures for the eyes, personnel with a medium or high workload could receive doses to the lens of the eye that would exceed the new annual equivalent dose limit, and could result in eye lens opacities over time.

(i) Several ophthalmological studies were conducted on a sample of interventional cardiologists and nurses who were attending cardiology congresses and who participated voluntarily under the coordination of the IAEA programme, the RELID study. Approximately 40–50% of interventionists and 20–40% of technicians or nurses, were found to have posterior subcapsular opacities compatible with injuries derived from exposure to ionising radiation. The incidence rate in interventionists was four to five times higher than that in unexposed individuals in the control group (approximately 40–50% vs 10%). Lifetime lens absorbed doses were estimated to reach several Gy in some cases.

(j) Extremity equivalent dose may be of concern, as the dose to the interventionalist’s hand that is nearest to the irradiated patient volume can be high and requires specific hand monitoring. Values for annual lower extremity equivalent doses up to 110 mSv have been found, despite the use of a protective curtain hanging on the side of the treatment couch. This exposure is attributed to the gap between the protective curtain and the floor, the size of which is dependent on the height of the x-ray table during exposure.

(o) Improved computational methodologies need to be developed to assess occupational doses, including equivalent dose to the lens of the eye, in high-dose procedures. These methods may be helpful to audit the regular and proper use of personal dosimeters and to assess the need for additional protection (e.g. protective glasses). Research programmes should pursue the development of computational technologies (not requiring dosimeters) together with personnel position sensing devices to assess personnel doses, including dose to the eye.

(r) Indirect approaches to dose assessment may be useful in identifying a lack of compliance in wearing personal dosimeters, and also in estimating occupational doses when personal dosimeters have not been used. These approaches may be based on area dosimetry of the scatter radiation near the patient (e.g. at the C-arm), together with conversion coefficients from patient-related quantities such as kerma-area product for different types of procedures and geometries to the dose to the lens of the eye of workers.

(u) The most important factor in protection of the head is the proper use of ceiling-suspended lead acrylic shields. They should always be included in interventional installations, as they can reduce doses to the whole head and neck by a factor of 2–10, depending on how efficiently they are positioned.

(v) Staff, such as nurses and anaesthesia personnel, who need to remain near the patient may benefit from the additional protection provided by movable (rolling) shields that can be positioned between them and the source of scattered radiation.

(w) As described in Point (h), under occupational exposure, the equivalent dose to the lens of the eye can exceed the new dose limit if protective measures are lacking. Over time, this could result in lens opacities. Conversely, if the interventional fluoroscopy equipment is operating correctly, procedure protocols have been optimised, the operator has been trained, and protective tools for the eyes are being used, the dose to the lens of the eye should be lower than the dose limit.

(x) A close fit of leaded glasses to the facial contours, particularly around the sides and underside of the glasses, is important because the clinician is looking at the image monitor during the x-ray exposures. As a result, the eyes may be irradiated from the side and below.

(y) Lead drapes attached to the bottom edge of the ceiling-suspended shield, as well as shielding drapes and pads, can be effective in protecting the hands in some procedures. This type of protection should be considered for procedures where the operator needs to be close to the source of scattered radiation (i.e. the irradiated volume of the patient). When placing disposable drapes on the patient, attention is required to avoid having the drapes in the primary beam, which might increase patient and operator exposure.

(z) Staff who stand near the patient table during interventions should be aware that the radiation field is more intense in the region adjacent to the beam entrance side. This is particularly important when projections are oblique or lateral. Doses to the head, upper body, and hands of the interventionalist from fluoroscopy with the tube positioned under the table will be substantially lower than the doses received by the lower extremities.

(aa) This is particularly true when no shielding curtains for the lower extremities are available, and when the table is at a higher position, so that the feet may stay unprotected even if the curtains are in place. Rolling lead shields, when available, decrease the effective dose to staff by more than 90% if used properly.

(bb) In summary, all staff in the room should wear protective aprons. Wraparound aprons are desirable for individuals who may not be able to face towards the patient at all times when the beam is on. The interventionalist should be protected by ceiling-suspended screens, table-suspended curtains, and shielding drapes when feasible. Staff can also reduce doses received during the use of high-dose acquisition modes (e.g. image acquisition series and digital subtraction angiography) and during injection of contrast media using an automatic injector by stepping back and increasing the distance to the patient. Staff, such as nurses and anaesthesia personnel, who need to remain near the patient can benefit from protection by movable screens. Other personnel should increase protection by increasing their distance from the irradiated volume of the patient or, if possible, leaving the room during image acquisition.

(dd) Current data do not justify precluding pregnant woman from performing interventions guided by radiological imaging completely if they follow proper procedures. Pregnancy, in any case, requires that the employer carefully reviews the exposure conditions and other aspects of occupational hazards (e.g. back pain with use of lead aprons) of the pregnant worker.

(14) CT-guided interventions can be performed by using intermittent CT scans while the physician steps behind a mobile shield or out of the scanner room, or by using CT fluoroscopy, with physician-controlled intermittent or continuous CT exposure during needle or device manipulation. CT fluoroscopy facilitates CT-guided biopsy procedures by allowing visualisation of the needle trajectory from skin entry to the target point. CT fluoroscopy is applicable to a wide variety of non-vascular interventions (Daly and Templeton, 1999). It is used for needle guidance during drainage of fluid collections and abscesses; spinal pain management; tumour ablation; and percutaneous needle biopsy in the neck, chest, spine, abdomen, and pelvis (Buls et al., 2003; Joemai et al., 2009; Hoang et al., 2011; Trumm et al., 2012). The principal advantage of CT fluoroscopy over standard CT is the ability to use real-time monitoring to access lesions that move within the body as a result of patient breathing or other motion. Its use allows interventions to be performed more rapidly and efficiently (Gianfelice et al., 2000b), and it is therefore popular. On the other hand, CT fluoroscopy also results in higher radiation doses to both the patient and the physician operator (Gianfelice et al., 2000a; Saidatul et al., 2010; Kim et al., 2011). As CT fluoroscopy images are noisier than conventional CT, this technique is predominantly used in cases of moving objects of high contrast, such as in lung biopsies.

(16) After catheterisation of the hepatic arteries, ⁹⁰Y microspheres (maximal beta-energy 2.27 MeV, half-life 64.1 h) are delivered under fluoroscopic control. Two types of ⁹⁰Y microspheres are used: resin microspheres (SIR-Spheres, SIRTEX, Lane Cove, Australia; diameter 20–60 µm) and glass microspheres (TheraSphere, Nordion, Ottawa, Ontario, Canada; diameter 22 µm). The rationale for SIRT is the dominant hepatic arterial supply of malignant lesions. SIRT has demonstrated a significant increase in patient survival time (Bester et al., 2012).

(17) SIRT is usually performed in two steps: in the first step, diagnostic angiography is combined with protective occlusion of non-target arteries. Then, shunting into the lung is estimated by means of a Single Photon Emission Computed Tomography (SPECT) scan of the lung and upper abdomen with ^99mTc-MAA particles injected into the hepatic artery. If lung shunting is <10%, SIRT with full ⁹⁰Y activity delivery is acceptable. A reduced amount of ⁹⁰Y activity (20–40% less) is recommended when shunting is 10–20% (SIRTEX). When shunting is >20%, SIRT is contraindicated. The second step, usually performed 1 or more days later, is catheterisation of the hepatic artery or other arteries supplying the hepatic tumours, and administration of the microspheres. Temporary balloon occlusion of non-target arteries and antireflux catheterisation during this second step are alternatives to protective occlusion prior to microsphere delivery (Hagspiel et al., 2013; Fischman et al., 2014).

(18) The second step includes dose calculation, preparation of the ⁹⁰Y spheres, and delivery via a catheter into the hepatic artery. Typical activities are 2–3 GBq for resin spheres (Jakobs et al., 2007) and 3–7 GBq for glass spheres (Andrews et al., 1994). Target dose is typically 120 Gy (range 80–150 Gy). Nuclide distribution may be examined either by planar or SPECT Bremsstrahlung imaging or PET-CT. PET-CT has higher spatial resolution, and quantification of delivered activity may be more accurate (Camacho et al., 2015).

(23) Annual effective doses incurred by staff depend on their function and role in the team (primary interventionalist, radiographer, nurse, anaesthesia provider), the number of interventions, the medical specifics and complexity of the cases, the patient population (e.g. paediatric patients, obese patients), and other factors such as the skill of the interventionalist and equipment as well as the use of fluoroscopic and cine times. In a review of the literature, Martin (2009) estimated that a case load of 500 cardiology procedures per year would result in an annual effective dose of approximately 2 mSv for the primary interventionalist. A maximum annual dose of 1.2 mSv [H_p(10) measured under apron] was observed for cardiologists at a Glasgow hospital (Martin, 2009). Other types of procedures that result in an effective dose per procedure greater than 10 µSv for the interventionalist might lead to annual effective doses as high as 10 mSv depending on whether thyroid shields are used. Lie et al. (2008) reported a maximum annual effective dose derived from combining the readings of two dosimeters, one under and one over the apron, of 11 mSv with a mean of 5 mSv. The outcome of a review of monthly effective doses (E), obtained during 2011 and 2012, performed by a dosimetry service provider in the USA is shown in Fig. 2.1. In total, 102,199 observations refer to workers monitored with two dosimeters (one over and one under the apron) and 196,526 observations refer to workers monitored with a single dosimeter located over the apron at the collar. The outcome revealed mean values of estimated effective dose of 0.12 and 0.27 mSv, with median values of 0.03 and 0.1 mSv, respectively (Yoder and Salasky, 2016).

(24) Sánchez et al. (2012) found monthly median under-apron doses of 0.11 mSv for cardiologists and <0.01 mSv for nurses in a study of 43 workers who conducted 1467 procedures. The over-apron doses were 0.4 mSv month⁻¹ for both cardiologists and nurses. The authors noted that perhaps as many as 50% of the cardiologists did not use their dosimeters correctly, often failing to wear the over-apron dosimeter (Sánchez et al., 2012). A multi-centre study on 39 physicians and nine assistants performing nine different types of procedures in 14 hospitals in Germany showed that the median body dose per procedure was 16 µSv for an unshielded person; the partial-body equivalent dose per procedure was 2.8 µSv to the lens of the eye, 4.1 µSv to the thyroid, 44 µSv to one of the feet, and 75 µSv to one of the hands. High exposures were measured to the hands, in some cases above the limit of 500 mSv (Häusler et al., 2009).

(25) As well as the primary interventionalist, other staff may also be subject to significant exposure, such as anaesthesia providers. Kong et al. (2015) showed that radiation exposure of anaesthesia providers not only depends on their workload, but largely varies with their positions and beam projections during interventional procedures. Beam projection accounts for a factor of 10 in effective dose and 200 in dose to the lens of the eye. A position close to the patient combined with left lateral projection causes higher exposure. Optimal arrangement of the anaesthesia device was found to be useful to reduce exposure.

(26) Data on occupational exposure from CT fluoroscopy guided interventions are limited. The highest doses are received by the physician’s hands, eyes, and thyroid (Saidatul et al., 2010). Use of thyroid shields provides substantial protection for the thyroid (Saidatul et al., 2010), which is especially important for younger professionals. Since average patient dose varies according to the type of procedure (Leng et al., 2011), average physician effective dose per case also varies according to the type of procedure, as would be expected; reported values measured over apron ranged from 2 to 25 μSv for H_p(10), with maximum values as high as 0.4 mSv per procedure (Paulson et al., 2001; Teeuwisse et al., 2001; Joemai et al., 2009). A variety of technical approaches and protection methods have been developed that can reduce occupational dose (Daly and Templeton, 1999; Paulson et al., 2001; Carlson et al., 2005; Hoang et al., 2011).

(27) The occupational radiation exposure from transcatheter aortic valve replacement or transcatheter aortic valve implantation depends on the approach (transfemoral or transapical). Values of H_p(10) up to 0.23 mSv in a single procedure were obtained by Shatila (2015) from the over-apron dosimeter of the primary operator (median value 0.11 mSv), as well as significant exposures to eight of 10 other workers.

(29) Most data on eye exposures are derived either from static experiments with phantoms or from individual monitors placed on the neck. A few studies have placed dosimeters closer to the eye on the forehead. Lie et al. (2008) compared thermoluminescence dosimeters (TLDs) placed near the left eye and between the eyes for 144 procedures, mainly cardiac. The median equivalent dose to the lens of the eye was observed to be 23 µSv per procedure, and the kerma-area product of the primary beam towards the patient was 0.4 µSv Gy⁻¹cm⁻². The left eye dose tended to be higher than that between the eyes due to the left eye being closer to the x-ray generator. Kicken et al. (1999) assessed the absorbed dose at the forehead for under-couch and over-couch x-ray systems. They found an average absorbed dose for the operator and assistant of 8 and 6 µGy per procedure, respectively, at one hospital; 16 and 14 µGy, respectively, at a second hospital; and 43 and 28 µGy, respectively, at a third hospital. The first two hospitals used an under-couch system and the third hospital used an over-couch x-ray tube that puts the head closer to the beam entrance to the patient-irradiated volume. Vañó et al. (2016) derived dose to the lens of the eye from dose measured over the apron, and compared dose to the lens of the eye in urologists per nephrolithotomy procedure with dose received by interventional cardiologists and radiologists. The report concluded that, due to the lack of protective shields in urology, the dose to urologists per procedure was 18.7 times higher than the dose received by interventional cardiologists who used ceiling-suspended shields.

(30) Within the European study on optimisation of radiological protection of medical personnel, TLD measurements and Monte Carlo simulation campaigns were performed for three cardiac and five interventional radiology procedures (Vanhavere et al., 2012). The selection was based on their potential impact on annual worker exposure (i.e. procedures with high frequency or high values of kerma-area product, or both). Operators were exposed substantially from embolisation procedures as well as from percutaneous transluminal angioplasty (PTA) of the lower limbs and renal arteries. During cerebral and carotid procedures, the doses to the operators were relatively low since femoral access is usually chosen and, therefore, the operator stands further away from the irradiated part of the patient compared with other procedures performed in the thoracic or abdominal region. Equivalent dose to the lens of the eye from digital subtraction angiography (DSA) and PTA was approximately 40 µSv, and the dose was up to 120 µSv for embolisations. Among the cardiac procedures included in the measurement campaign, higher operator doses were delivered from the implantation of pacemakers and cardiac defibrillators, despite their relatively low kerma-area product values; this is due to the fact that fluoroscopy alone is used in these interventions. The reason for the higher occupational doses from these procedures is that operators work very close to the irradiation field, and often work without any protective shielding. Average eye doses lie within the range of 40–60 μSv.

(31) Other studies indicate that annual equivalent dose to the eyes of some interventional clinicians may be in the region of 50–100 mSv (Vañó et al., 2008a; Ciraj-Bjelac et al., 2010; Thornton et al., 2010; Koukorava et al., 2011; Jacob et al., 2013; Martin and Magee, 2013; IAEA, 2014b; Principi et al., 2015). Thus, radiation dose to the lens of the eye for interventional clinicians with high workloads can readily exceed the revised 20 mSv dose limit for the lens of the eye (ICRP, 2012), unless appropriate radiological protection measures are put in place.

(33) In the study on Optimization of Radiation Protection for Medical Staff (ORAMED), an average equivalent dose per procedure to the left hand of approximately 240 μSv was observed for DSA/PTA of the lower limbs, approximately 320 μSv for embolisations, and approximately 60 μSv for cerebral DSA/PTA procedures. Average doses of 410 μSv have been recorded for the left finger for cardioverter defibrillator implantation, while for cardiac angiography/angioplasty and radiofrequency ablations, the respective values were 180 μSv and 60 μSv (Vanhavere et al., 2012).

(34) Felmlee et al. (1991) made scatter measurements at various distances from a 12-cm × 15-cm field with a phantom entrance absorbed dose rate of approximately 65 mGy min⁻¹ and exit dose rate of 0.7 mGy min⁻¹. The scatter dose rates in the lateral direction ranged from 0.7 mGy min⁻¹ at a distance of 0 cm, to 0.35 mGy min⁻¹ at a distance of 5 cm, and 0.13 mGy min⁻¹ at a distance of 15 cm (Felmlee et al., 1991). The variation in reported hand doses is explained by the large dose gradients near the x-ray beam, movement and placement of the hands, and whether the interventional procedure involves femoral, percutaneous, or internal jugular vein catheter insertion that places the physician in different positions relative to the patient and x-ray tube (Whitby and Martin, 2005; Martin, 2009). Hand doses also tend to be much larger for over-table x-ray units due to the greater scatter from the primary beam.

(35) Poor technique in CT fluoroscopy can result in the physician’s hands being placed in the direct beam (Buls et al., 2003), reaching the annual dose limit of 500 mSv in a few minutes.

(37) The group found that, without protection, the lower limb dose was frequently greater than the hand dose, with a mean leg dose between 0.19 and 2.61 mSv per procedure without any protection, and between 0.02 and 0.5 mSv per procedure with a protective curtain (Artschan et al., 2014). The ORAMED study showed leg doses of 160–250 μSv (Vanhavere et al., 2012).

Interventional radiologists and other staff are exposed during transcatheter delivery into the hepatic artery.

Nurses are exposed after the procedure until patient discharge.

(39) Only a few papers on occupational doses from SIRT have been published. Occupational exposure from SIRT procedures is caused by x rays with relatively low dose rates and by direct beta radiation, especially to the hands and fingers, with high dose rates if precautions are inadequate. In addition to the dose to the hands of workers preparing the individual patient dose and to the physician implanting the microspheres, there is potential for significant contamination hazard. Specific advice to reduce this hazard is given in Section 5. Exposure data are 43.5 mSv MBq⁻¹ h⁻¹ skin equivalent dose due to contact with a 5-mL syringe and 1.35 mSv kBq⁻¹ h⁻¹ due to contamination with 50 μL on 1 cm² (Kemerink et al., 2012).

(41) PET-CT-guided biopsies are not common. They are performed when CT alone is not sufficient to identify the area of possible cancer (Werner et al., 2011; Aparici and Win, 2014). PET-CT-guided interventional procedures typically use ¹⁸F-FDG. Ryan et al. (2013b) quantified occupational radiation exposure and found a median effective dose per procedure of 0.02 (range 0–0.13) mSv for the primary operator, 0.01 (range 0–0.05) mSv for the nurse and anaesthesia provider, and 0.02 (range 0–0.5) mSv for the radiographer. The median extremity equivalent dose for the operator was 0.05 (range 0–0.62) mSv per procedure. Radiation exposure of the workers correlated with the duration of the procedure, and with the use of in-room image guidance. The authors concluded that operator effective dose from PET-CT-guided procedures was not significantly different from typical doses from fluoroscopically guided procedures. The major determinant of radiation exposure to the operator from PET-CT-guided interventional procedures is time spent in close proximity to the patient. As novel PET isotopes are developed, they may result in different dose profiles near the patient (Holland et al., 2010; Williamson and Dauer, 2014).

(42) With regard to fingertip doses from ¹⁸F-FDG, Sánchez et al. (2015) measured dose reductions from using a full automatic system for preparing and infusing the FDG. The results show a reduction in the average skin dose to the fingertips of radiographers from 223 to 83 µSv GBq⁻¹ (63%) from preparing the radiopharmaceutical. The average skin dose to the fingertips of nurses was reduced from 83 to 11 µSv GBq⁻¹ (87%) from infusion to the patient. The accuracy of the delivered activity was 2%.

(44) One of the first reported cases of radiation-induced opacities in interventional radiologists was in 1998, and the reason for the radiation injuries was the use of a non-optimised interventional radiology laboratory and the lack of a radiological protection programme (Vañó et al., 1998). In 2004, Haskal presented the results of a pilot study of x-ray-associated lens changes in 59 practising interventional radiologists; 37% of those screened had detectable posterior lens changes consistent with radiation exposure (Haskal, 2004; Junk et al., 2004). Although lens radiation doses were not reported, the authors noted that the frequency and severity of posterior subcapsular lens opacities increased as a function of age and years of practice, thus suggesting a possible dose–effect relationship.

(45) Following these findings, in 2008, IAEA promoted a project called ‘Retrospective Evaluation of Lens Injuries and Dose’ (RELID) for interventional cardiology (IAEA, 2016), with the objectives of estimating occupational lens doses and evaluating possible lens opacities.

(46) Since no personal dosimetry data were available, occupational lens doses were estimated in most cases by combining published typical scatter dose values (Vañó et al., 2008a,b) with information on the declared numbers of working years, workload, fluoroscopy and cine exposure conditions, radiological equipment used, location of the worker in the room, and use of radiological protection tools. Availability of some personal monitoring badge data helped in assessing the correlation.

(47) For the ophthalmological examination of posterior subcapsular opacities, Merriam–Focht scores were used (Ciraj-Bjelac et al., 2010, 2012; Rehani et al., 2011; Vañó et al., 2010, 2013a). The scoring (i.e. 0.5, 1.0, 1.5, etc.) is done separately for each eye. In total, eight surveys were performed under the RELID study (Bogotá 2008, Kuala Lumpur 2009, Montevideo 2009, Varna 2009, Sofia 2009, Bangkok 2009, Buenos Aires 2010, and Kuala Lumpur 2011).

(48) The RELID study concluded that workers in cardiac catheterisation laboratories show an increased prevalence of eye lens opacities when professionals have been working for several years without the proper use of radiological protection tools. Approximately 40–50% of interventionists and 20–40% of technicians or nurses, voluntarily attending the lens injury examination (during cardiology congresses), were found to have posterior subcapsular opacities compatible with injuries derived from exposure to ionising radiation. The incidence rate in interventionists was four to five times higher than that of the unexposed individuals in the control group (approximately 40–50% vs 10%). Estimated lens doses reached several Gy in some cases during a full professional life. However, it is still not clear if lens opacities progress to visually disabling cataracts.

(49) Although a radiation-induced decrease in contrast sensitivity has not been reported in the study populations, in the last RELID study (Vañó et al., 2013a), a restricted contrast sensitivity test was performed for approximately 20% of the participants with observable lens changes on slit lamp examination. The contrast sensitivity curve for these participants resulted in a significant loss of contrast in comparison with the standardised normal data. Retrospective dose estimations are necessary to look for correlations between radiation dose and lens opacities (Vañó et al., 2013a). Comprehensive reviews of radiation effects on the lens of the eye are provided in ICRP and NCRP publications (ICRP, 2012; NCRP, 2016).

(50) In many of these studies, there was irregular use of personal dosimeters and protective tools. These results point to the need for improving radiological protection, following the recommendations given in Section 5.

(57) Similarly, an important finding in ophthalmological studies (RELID study) is irregular use of personal dosimeters, and poor adherence to the ICRP recommendation to use two dosimeters, with one dosimeter located at collar level over the apron from which dose to the lens of the eye can be inferred. A study by Vañó et al. (2013b) showed that only approximately 50% of the interventionalists reported that they use personal dosimeters, and only 30% reported their use on a regular basis. Approximately 90% of nurses and technicians reported the use of personal dosimeters, but regular use was only reported by approximately 40%. Even when used, dosimeters were worn under the apron in most cases, so any retrospective evaluation of ocular radiation dose using these devices is likely to be inaccurate. In a previous study, Niklason et al. (1993) showed that half of the workers did not use their personal dosimeters regularly.

(58) A retrospective study of 15 years follow-up in a cardiology department observed that 20–30% of cardiologists were not using their dosimeters routinely (Vañó et al., 2006). In surveys conducted by IAEA during various radiological protection training courses, which included cardiologists from over 56 countries, responses indicated that 33–77% of interventional cardiologists used radiation badges routinely (IAEA, 2014b).

(59) If two dosimeters, meant for placement under and over the apron, show similar readings, this indicates that their placement may have been randomly reversed. Another disparity can arise when protective glasses are only used for some procedures. Therefore, a consistent deployment of dosimeters together with consistent use of the protective devices needs to be stressed. Workers need one set of instructions on how many dosimeters to use and where to place them which is specific to their most restrictive duty or risk of exposure. The Commission (ICRP, 2000b) and others (NCRP, 2010) recommend that interventional radiology departments should develop a policy and good habits for workers to wear two dosimeters.

(67) For occupationally exposed workers in medical interventional procedures, the dose limits for workers recommended by ICRP apply. In planned exposure situations, recommended dose limits for workers were established in Publication 103 (ICRP, 2007a), with an updated limit for the lens of the eye in the ICRP statement on tissue reactions (ICRP, 2012).

(68) The following limits apply:

Whole body: an effective dose of 20 mSv year⁻¹, averaged over defined periods of 5 years, provided that the effective dose does not exceed 50 mSv in any single year.

Extremities: hands and feet, an equivalent dose of 500 mSv year⁻¹.

Skin: an equivalent dose of 500 mSv year⁻¹, averaged over 1-cm² area of skin regardless of the area exposed.

Lens of the eye: an equivalent dose limit for the lens of the eye of 20 mSv year⁻¹, averaged over defined periods of 5 years, provided that the equivalent dose to the lens of the eye does not exceed 50 mSv in any single year.

(69) The objective of the recommended limits is to ensure that the occurrence of stochastic health effects is kept below unacceptable levels, and that tissue reactions (deterministic effects) are avoided.

(72) Exceeding a monthly investigation level provides an alert that protection was less than optimal in that period of time, and a review of existing radiological protection is needed. The increase in the dosimeter reading may be due to a substantial increase in the number of interventions, or in the dose per procedure, which may be due to an increase in procedure complexity or to a decrease in compliance with protection measures.

(73) In 2000, the World Health Organization (WHO) recommended that an investigation should be carried out when monthly exposure reaches 0.5 mSv for effective dose, 5 mSv for dose to the lens of the eye, or 15 mSv to the hands or extremities (WHO, 2000). Following the new annual limit of equivalent dose to the lens of the eye, the investigation levels should be lowered accordingly. An investigation level of 2 mSv month⁻¹, using the reading from the collar dosimeter, has been proposed for interventional cardiologists (Durán et al., 2013, endorsed by the Pacific Society of Interventional Cardiology, the European Association of Percutaneous Cardiovascular Interventions, the Latin American Society of Interventional Cardiology, and the Society for Cardiovascular Angiography and Interventions).

(74) An investigation level in terms of a monthly dose should be such that when extrapolated to a year, it would not exceed the relevant dose limit or dose constraint. In addition, as described in Section 2, personal dosimeters are not always worn or are worn improperly (Padovani et al., 2011; Sánchez et al., 2012). Investigation levels can be helpful in this situation by establishing minimum dose values for the over-apron and hand dosimeters, thus providing an alert for possible poor compliance with procedures for wearing dosimeters.

(80) Occupational exposure rests on a series of assumptions regarding the relationship between what is measured by a dosimeter and the dose received by an individual. Standards include accuracy requirements and uncertainties of the dosimetry system, so that these assumptions hold for the relationship between operational and protection quantities. Ensuring that workers wear the dosimeters correctly during all work time is the most important component of this series of assumptions and relationships. No dose to an individual can be estimated reasonably in highly variable radiation fields without having some type of individual monitoring on the workers during all times of exposure. Auditing compliance with procedures is important to verify that the workers wear the dosimeters regularly and correctly.

(83) For monitoring of the hands, small ring-shaped dosimeters are used due to their relative ease of fit under surgical gloves. Rings can be sized for different finger diameters, but attention is required to the fact that fingers may swell during long procedures. In addition, some additional features are important such as sterilisation capability and low interference with tactile sensation in the operator’s ability to manoeuvre catheters and instruments precisely. Fingertip sachets that fit over a finger have been used as an alternative to ring dosimeters, and are placed with the radiation sensor at the most proximal part of the hand where the largest doses may occur. An alternative solution that reduces interference with tactile sensation consists of wearing a TLD-type dosimeter on a finger nearest to the irradiated volume of the patient.

(84) The physical construction of the dosimeter has to be compatible with the intended wearing location. Infection control is a particular concern for ring dosimeters because they may not withstand a sterilisation process, and they are typically worn during procedures where infection control is essential.

(85) Dosimeters worn on the body should not create sharp pressure points that cause discomfort when placed between the heavy leaded apron and the user’s clothing. If whole-body dosimeters are placed near the neck on top of the leaded apron or over a protective thyroid shield to assess doses to unshielded areas, they should not have any edges that could irritate the neck or chin area. All methods of attachment should be strong enough to prevent dislodging during strenuous use, but not cause dislocation of protective aprons or damage to clothing in the event the dosimeter catches on a foreign object.

(86) Active personal dosimeters (APDs) or electronic dosimeters are used for optimisation monitoring, or for special studies that require analysis of dose by procedure or discern aspects of a procedure (e.g. the relative dose received during fluoroscopy compared with image acquisition). APDs are able to provide immediate information about dose rate, so rapid feedback is available to staff against which they can assess changes to their behaviour that result in lower dose rates and, subsequently, lower accumulated doses. Dose rate information is needed if actions are desired within a procedure, as it can lead directly to procedural change. In addition, APDs provide information on the time of each exposure, which facilitates correlation of occupational and patient exposures, and auditing of the wearing of the APD during interventions.

(87) Electronic dosimeters are useful for educating staff. For example, the large dose rate reduction when a ceiling-suspended shield is brought into place is very illustrative, and encourages clinicians to use these shields routinely.

(88) Until recently, electronic dosimeters were too large to use conveniently under leaded aprons. The power requirements of older units added to the operational overhead of assuring charged batteries so that the units would not power off during a lengthy procedure. Advances in power management and wireless transmission of signals from an electronic dosimeter to a base station have overcome some of the disadvantages of using electronic dosimeters. Some manufacturers of interventional systems have included electronic monitors with wireless data transmission, so dose rates and doses can be viewed adjacent to or as part of the image video screens that the operator uses.

(89) Optimisation monitoring does not strickly need to conform to the dose quantities required for compliance monitoring. Optimisation seeks to compare relative values resulting from changes in conditions, in order to evaluate effectiveness of various actions to reduce dose. Electronic dosimeters are usually calibrated to assess operational quantities without taking into account the non-uniform irradiation of the body during interventional procedures. They indicate the dose at a single point and make no inferences regarding effective doses or doses at some distance from the dosimeter. Conceptually, there is no technical reason why multiple electronic dosimeters could not be worn and the data combined to yield compliance-type dose information, but practical issues tend to limit the use of electronic dosimeters to investigatory and optimisation monitoring.

(90) Electronic dosimeters have not been developed for routine use on the fingers or near the eyes. An attempt was made to place small electronic sensors on the fingers, but the electrical cables leading back to the power source and electronics were not convenient or practical for routine use.

(133) The ulnar aspect of the hand, which is side-on to the x-ray beam and closer to the irradiated volume of the patient, receives a higher dose, so dosimeters should be worn either on the little finger or the side of the wrist closest to the x-ray tube (Whitby and Martin, 2005; Vanhavere et al., 2012).

(134) Proper dosimetry to evaluate dose to the hands and fingers is not easy in clinical practice. A method to estimate hand dose in interventional radiology is a wrist dosimeter but, due to the inhomogeneity of the radiation field near the patient and the potential of introducing part of the hands into the direct beam, the dose measured by wrist dosimeters could be much lower than the actual finger dose.

(135) There are specific challenges in the assessment of finger dose in CT fluoroscopy. A primary beam from a CT scanner can deliver doses of hundreds of µGy in a few seconds. As indicated in Section 2, if an operator's hand is placed in the direct beam, the annual dose limit of 500 mSv could be reached in a few minutes (Buls et al., 2003; Saidatul et al., 2010). Due to the relatively narrow CT beams, of the order of 5-mm wide, a dosimeter worn at the base of the finger may fail to record the exposure of the rest of the finger to the primary beam, and in most cases, only scattered radiation would be measured. Extremity monitoring in CT fluoroscopy is therefore of limited value in this case, and may provide false reassurance.

(136) The CT images should alert an operator if part of their hand enters the primary beam. In addition, the capability to review CT images provides a means of controlling these exposures and facilitating investigations. A dose assessment can be made through measurements in the actual positions where exposures have occurred for appropriate lengths of time. Since dose levels in the CT beam can be measured, dose data should also be included in the assessments.

(144) Changing beam projection angle to avoid exposing the same skin area all the time may avoid patient skin injuries in complex and long interventions, but the way it affects the staff exposure depends on the extent of gantry angulation and the position of the x-ray tube with respect to the position of the interventionalist. Staff dose from scattered radiation increases when the x-ray tube is on the same side as the interventionalist with respect to the irradiated volume of the patient.

(150) A number of professional societies, radiological protection organisations, and others have issued guidelines on practices to be followed, and made recommendations on the use of protective devices (Miller et al., 2010; NCRP, 2010; Chambers et al., 2011; Sauren et al., 2011; Durán et al., 2013; ICRP, 2013a,b; Hiles et al., 2016).

(152) The fit of the protective apron is often more important to reduce effective dose than the lead equivalence of the apron (Detorie et al., 2007). Thoracic organs, including the lungs and oesophagus, may receive higher levels of exposure when the operator is irradiated from the side through the armholes of the apron (Franken, 2002). This is particularly important for protection of the axillary portion of the female breast. Some facilities favour aprons with shoulder protection to reduce this exposure (Guersen et al., 2013).

(154) There are lighter-weight aprons containing composite layers or bi-layers of high atomic number elements, such as tin or bismuth, as well as, or instead of, lead. Similar attenuation can be achieved with lighter-weight aprons because the alternative metals are more efficient per unit mass than lead for absorbing x-ray photons with energies between 40 keV and 88 keV. These aprons may be more effective for attenuating scattered x rays from tube voltages of 70–80 kV, but less effective for tube voltages above 100 kV (Christodoulou et al., 2003). They provide a reasonable alternative where weight reduction is required to alleviate back or neck problems. Lightweight or ‘lead-free’ aprons have a different x-ray transmission from aprons containing lead for different x-ray spectra.

(155) Manufacturers often specify the attenuation properties in terms of lead equivalence (e.g. 0.5-mm lead equivalent); these data without further qualification can be misleading (Finnerty and Brennan, 2005; Schlattl et al., 2007; Eder et al., 2010) since attenuation varies significantly over the photon energy spectrum, with the largest variations occurring in the diagnostic imaging range. Manufacturers might also have tested the transmission of the protective garment in narrow-beam geometry using primary radiation, rather than broad-beam geometry using scattered radiation which is the situation that staff will be in. One study concluded that these aprons provide less lead equivalent thickness than that stated on their labels and their manufacturing certificates (Papadopoulos et al., 2009). Users and patients who use lead-free x-ray protective clothing might unknowingly be exposed to a greater dose than generally assumed.

(156) The indication of the protective value of garments should be accompanied by specification of the characteristics of the radiation beams (IEC, 2014) and the geometries used to measure the attenuation, and the combination of measurements made at different beam qualities to reflect the conditions under which the garment is used. Jones and Wagner (2013) concluded that measurement of attenuation or lead equivalence at a single beam quality is inadequate, and suggested that specifications of the protective value of protective garments should accurately reflect the protection provided across the range of beam qualities for which it will be used, and these specifications should be relatively easy to verify by the end-users. They further suggest four standard beam qualities that should be selected, that penetration through protective garments should be measured for these beam qualities with the garment configured as it would be for clinical use, and that the protection index should be weighted by the distribution of workload with respect to the four beam qualities (Jones and Wagner, 2013).

(157) It is important for users to understand the labelling of the lead equivalence for the particular protective garment. For example, where a manufacturer quotes ‘0.35-mm Pb’, it could be 0.35-mm Pb only when two flaps are overlapped, rather than 0.35 mm Pb through every part of the garment; users might falsely believe that when the flaps overlap, they get 0.7-mm Pb equivalence.

(166) The protection to the eyes provided by ceiling-suspended shields or lead glasses can be quantified in terms of dose reduction factors (DRFs). Reports on dose reductions to the eyes achieved through use of ceiling-suspended shields give varying DRFs. A large-scale report of clinical measurements for interventional procedures gave DRFs between 1.3 and 7 (Vanhavere et al., 2012). A review comparing doses from groups at different centres performing similar procedures gave DRFs between 0.7 and 2.5 (Jacob et al., 2013), and a study comparing dose rates for periods when radiologists were using and not using shields gave a DRF of 5 when the shield was in use (Magee et al., 2014). However, phantom simulations with precise positioning of the ceiling-suspended shield yield higher values. In a phantom study, Galster et al. (2013) reported DRF values between 8.5 and 17.6 for transjugular intrahepatic portosystemic shunt (TIPSS) creation, embolisation of abdominal haemorrhage, and pelvic embolisation. Ceiling-suspended shields demonstrated a significantly higher dose reduction than lead glasses, and protect the entire head and neck, not just the eyes (Galster et al., 2013). One clinical study with careful placement of a shield for percutaneous coronary interventions observed a DRF of 19 (Maeder et al., 2006).

(167) When use of a ceiling-suspended shield is possible, the level of dose reduction achieved depends on the use of the shield and how effectively it is positioned. The shield should be placed just above the patient, with the operator viewing the irradiated area of the patient through the shield. This is an important element of radiological protection training for interventionalists (Vanhavere et al., 2012). However, it is often difficult to use these shields effectively with the x-ray tube in lateral or oblique projections. Effective use of shields requires their continual repositioning as the x-ray tube and couch are moved. Thus, although the shields give good protection in principle, difficulties in their effective deployment for the range of projections used in clinical procedures may limit the overall level of protection achievable in routine use. Nonetheless, with diligence, DRFs of 2–5 should be achievable. This reduction should allow interventionalists to keep doses to the eye below the limit, and avoid eye lens opacities which may otherwise occur through the accumulation of dose over a professional working life.

(168) Vañó et al. (2015b) estimated that more than 800 procedures per year and per operator would be needed to reach the new dose limits for the lens of the eye for three interventional specialties (cardiology, neurology, and radiology) using the conservative approach of estimating dose to the lens of the eye from the over-apron collar dosimeter, and assuming proper use of ceiling-suspended protective shields (Vañó et al., 2015b).

(169) Some authors have shown attenuation and potential dose saving to the brain of the interventionalists (Alazzoni et al., 2015; Uthoff et al., 2013) provided by radiation-absorbing surgical caps, while others have focused on measurements of dose distributions inside the head using radiochromic films on a RANDO phantom, and have reported that the radiation absorbing surgical cap provides essentially no protection to the brain of an interventionalist (Fetterly et al., 2017). The authors have explained that this is due to the geometry of the irradiation (i.e. the fact that the scatter originates from a location inferior to the interventionalist’s head, and the cap cannot protect the brain from radiation coming from below).

(170) This finding is consistent with the fact that actual doses to staff depend on a number of factors, and radiological protection needs to be optimised in a systematic approach by knowing and taking care of all of them, including the effective use of ceiling-suspended shields. The first consideration should be to find out if there is a need to provide additional brain shielding, given that the published data demonstrate no evidence of an increased incidence of brain cancers from radiation doses to the brain of interventional staff, as discussed in Section 2. Staff should have access to consultation with a medical physicist or an expert on radiological protection familiar with interventional procedures and questions related to radiological protection, including the assessment of brain radiation exposure, before undertaking additional protection measures.

(173) DRF values between 5 and 10 have been reported from experimental measurements for a variety of lead glasses when protecting against x rays incident from the front in the same horizontal plane as the eyes (Moore et al., 1980; Marshall et al., 1992; Thornton et al., 2010; McVey et al., 2013; Van Rooijen et al., 2014) and Monte Carlo simulations (Carinou et al., 2011; Koukorava et al., 2014).

(174) However, in clinical practice, the DRF depends on the angle at which scattered radiation from the patient reaches the head (McVey et al., 2013; Magee et al., 2014; Van Rooijen et al., 2014), and often the interventionalists' eyes are irradiated from the side and below, at an angle between 20° and 90° with the horizontal plane. DRF values tend to be lower for irradiation from the side due to the larger spaces left between the glasses and the head for the prescription spectacles (Magee et al., 2014). For the majority of the time that interventionalists carry out a procedure when x rays are being emitted, they will be viewing the images on the monitor rather than looking towards the patient. Then, scattered radiation may be able to pass through gaps behind the lenses and through parts of the frame that are not protected to irradiate the eyes directly.

(175) Also, unattenuated x rays incident on tissues that are close to the eyes and scattered from these tissues become a major source of exposure to the lens of the eye when protective eyewear is worn (Moore et al., 1980; Cousin et al., 1987; Marshall et al., 1992; McVey et al., 2013; Koukorava et al., 2014; Magee et al., 2014). For exposures from the front, differences between various categories of glasses relate to the sizes of the lenses, and the proximity of unprotected and therefore irradiated tissue. For exposures from the side, the eye dose depends on the closeness of fit to the facial contours and the extent of protection from the side. When the radiation scattered by the patient is incident towards the eye from below, it may enter directly through the gaps underneath the lenses of the glasses, without any additional scattering.

(176) The majority of lead glasses have a protection equivalent to 0.75 mm or 0.5 mm of lead, and many have protection in side shields of 0.5-mm or 0.3-mm lead equivalence. The designs can be divided into a number of categories which are listed below:

(A) Purpose-designed lead glasses with large flat lenses and protective side shields.

(B) Wraparound lead glasses with front lenses angled to provide more protection for radiation incident from the side.

(C) Lead glasses adapted from conventional spectacles with lead glass side shields added.

(D) ‘Fit-over’ glasses, similar in design to Category (A), but arranged to fit over conventional spectacles.

(E) Facemasks of lower lead equivalence, held in place by a headband.

(177) Custom-designed lead glasses of Categories (A) and (B), having a lead equivalence of 0.75 mm, provide protection for the eyes with DRFs between 3.5 and 6 (Cousin et al., 1987; Vanhavere et al., 2012; Koukorava et al., 2014; Magee et al., 2014; Principi et al., 2015; Martin, 2016), and 0.50-mm lead equivalence glasses might provide DRFs of 3 to 4. Wraparound lead glasses provide better protection for radiation incident from the side and below because the gaps between the frames and head tend to be smaller. Glasses based on adaptations of standard spectacles of 0.75-mm lead with added side shields have DRFs between 3 and 4, as gaps between the glasses and the head tend to be larger (Magee et al., 2014).

(178) ‘Fit-over’ glasses designed to be worn over prescription spectacles are bulky, and have larger gaps underneath to allow wearing of conventional spectacles. Verification that critical parts of the frames are of protective material is important, as some models, particularly the heavier ‘fit-over’ glasses, do not use protection in the frames in order to keep the weight down.

(179) For head positions behind a ceiling-suspended screen, Galster et al. (2013) reported additional DRFs for lead glasses between 1.8 and 5.8.

(180) Facemasks or visors of lower lead equivalence such as 0.1 mm cover the whole of the face, and so reduce the exposure of regions of the head surrounding the eyes that would make a significant contribution to the dose to the eye from backscatter (Martin, 2016). Despite the lower lead equivalence, they provide a viable alternative to lead glasses, but are sometimes not favoured by clinicians due to their size and their tendency to fog.

(181) Measurements of the protection offered by lead glasses can provide useful data from which adjustments to dosimeter reading values recorded by unshielded eye dosimeters can be made to derive a dose representing that to the lens of the eye for any interventionalists for whom it could be guaranteed that they wore the protective eyewear consistently. However, any calculations assume that lead glasses are worn for every procedure. Therefore, for an attenuation factor to be applied, quality controls should be in place with regular documented checks to confirm that the interventionalist concerned always wears the protective eyewear.

(182) The factor applied could be one based on measurements with the glasses concerned, but should take account of exposure from x rays at angles encountered in clinical practice. The measurement technique and the results should be documented, and the DRF applied should not be greater than 4.

(183) Where no measurements are available to confirm the DRF, but the glasses are of Category (A) or (B) and incorporate the equivalent of at least 0.5 mm of lead, a DRF of 2 represents a conservative approach to account for the protection offered by the glasses (Magee et al., 2014).

(184) The protection afforded to the contralateral eye by almost all models of lead glasses is lower than that to the eye adjacent to the x-ray source, with DRFs for the contralateral eye being between 1.5 and 2 (Geber et al., 2011; Galster et al., 2013; Van Rooijen et al., 2014; Fetterly et al., 2017). This is due to penetration of radiation through the gap corresponding to the nasal cavity, when the head is at an angle to the direction of the beam, and because a substantial portion of the dose is due to scatter from within the surrounding tissues. However, the dose to the lens of the contralateral eye is still less than that to the eye adjacent to the source.

(185) In summary, a variety of lead glasses are available, but care should be taken in their selection. A close fit to facial contours, particularly around the underside of the glasses, can be more important than lead equivalence, as the glasses should also provide protection against exposures from below and from the side.

(187) Thornton et al. (2010) evaluated the impact of common radiation-shielding strategies, used alone and in combination, on scattered dose to the fluoroscopy operator’s eyes. Operator phantom lens radiation dose rate was recorded with and without a leaded table skirt, non-leaded and leaded (0.75-mm lead equivalent) eye glasses, disposable tungsten-antimony drapes (0.25-mm lead equivalent), and suspended (0.5-mm lead equivalent) transparent leaded shields. Lens dose measurements were also obtained in right and left 15° anterior obliquities with the operator at the upper abdomen, and during DSA (two images per second) with the operator at the patient’s groin. Each strategy’s shielding efficacy was expressed as a reduction factor of the lens dose rate compared with the unshielded condition. Use of leaded glasses alone reduced the lens dose rate by a factor of 5–10; and scatter-shielding drapes alone reduced the dose rate by a factor of 5–25. Use of both implemented together always provided more protection than either used alone, reducing the dose rate by a factor of 25 or more (Thornton et al., 2010).

(189) Employing different access routes for the intervention influences the positions of the operator's hands during procedures and has a substantial effect on the dose level (Fig. 5.3). For cardiologists, introduction of catheters via the radial artery rather than the femoral artery has advantages in achieving patient mobility more quickly, but the cardiologists’ hands are closer to the x-ray beam and so the doses they receive, particularly to the side of the hand, are higher (Mann et al., 1996).

(190) In interventional radiology, femoral access is used much of the time, but percutaneous procedures such as percutaneous biliary drainage, nephrostomy tube placement, and gastrostomy placement require the operator to manipulate catheters inserted close to the area being imaged, and thus can give relatively high doses to the finger tips (Whitby and Martin, 2005).

(191) In procedures such as TIPSS, in which the radiologist gains access via the internal jugular vein, the hands are located further from the area being imaged, but TIPSS procedures can be technically challenging, fluoroscopy times are long, and doses are relatively high (Fig. 5.4).

(192) The hand that holds the catheter is usually closer to the edge of the x-ray beam and receives the higher dose, while the other hand performs the manipulations (Figs 5.5 and 5.6).

(193) Ceiling-suspended shields provide good protection for the head and upper body, but the hands are generally positioned below the shield and so receive less protection. However, some reduction can be achieved with careful practices (Maeder et al., 2006). Lead/rubber drapes attached to the bottom edge of the shield can be effective in protecting the hands for some procedures (Vanhavere et al., 2012).

(194) Freestanding adjustable over-table shields can shield the operator’s hands, but the hands may stretch underneath the shield and so receive less protection. Protective drapes and pads can also offer good protection for the hands, and have been shown to achieve a 29-fold reduction in the dose to the hands in one study (King et al., 2002).

(195) Thin protective gloves are available, but reports of the protection offered are varied (15–60%). If a hand protected by a glove strays into the x-ray field, the dose rate will be increased automatically to compensate for the attenuation, thus increasing patient exposure without achieving any protection of the hand of the interventionalist (Wagner and Mulhern, 1996). New shielding materials (e.g. bismuth) have also been proposed as a hand cream for hand protection, subsequently to be covered with a surgical glove to provide containment of the cream material (McCaffrey et al., 2012). This cream has the same potential to increase dose if the hand is placed in the x-ray field. Moreover, the reduction in tactile feedback from radiation-attenuating material may lead to an increase in fluoroscopy time or CT exposure time for delicate procedures (NCRP, 2010).

(197) Lead curtains attached to the side of the couch, typically with a lead equivalence of 0.5 mm, provide the operator with the best protection (Whitby and Martin, 2003; Shortt et al., 2007). These drapes can reduce doses to the legs by factors of 10–20 if positioned correctly throughout a procedure (Martin, 2009), but factors between 2 and 7 are typical in practice (Vanhavere et al., 2012). Such drapes should be specified for all interventional facilities.

(198) A lead curtain that is attached to the table and hangs down from it has the advantage of being as close as possible to the source, and is always in place so no conscious decision is needed to use it. For the majority of procedures, where the interventionalist stands at the side of the table, a lead drape attached to the table provides a good option. However, it rarely protects the feet fully.

(199) Usually the leaded curtain attached to the table does not extend for the full length of the table, so positioning is important for protection of both the operator and assistants. Operators standing at the side of the table will be adequately protected, but when an interventionalist stands at or near the head of the table, as in the case of TIPSS procedures, the drape will only provide protection for the interventionalist if it can be moved to the head of the table. These shields may be less effective for procedures in which the interventionalist is positioned near the head or foot of the table. For such procedures, other staff may need to stand to the side of the table, and they will require leg protection.

(200) Mobile freestanding shields are available for protecting the legs. A conscious decision needs to be made to put them in place before the start of the procedure in order to preserve a sterile environment. There is a risk of collision with the couch and C-arm when either is moved up and down, tilted or angled. Such shields may also be used for protecting other staff who are assisting with procedures.

(201) Stepping back from the couch during radiography is an effective method to reduce occupational dose; this is rarely possible during fluoroscopy, as the operator must be close enough to the patient to perform the procedure, but it is possible during acquisition series and during injection of contrast media using an automatic injector.