This Article Abstract Full Text (PDF) Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Dendy, P P Search for Related Content PubMed PubMed Citation Articles by Dendy, P P

Radiation risks in interventional radiology

1A Coppice Avenue, Great Shelford, Cambridge CB22 5AQ, UK

Correspondence: P P Dendy, 1A Coppice Avenue, Great Shelford, Cambridge CB22 5AQ, UK. E-mail: dendy{at}waitrose.com

	Abstract

Top Abstract Introduction Current trends in interventional... Doses to patients Staff doses Equipment factors and programme... Education and training Conclusions and future work References

 
The number, diversity and complexity of interventional radiological examinations have all increased markedly in recent years, and it is widely recognized that some of these procedures carry greater risks than many other radiological procedures. This Commentary uses a meeting on "Radiation Protection in Interventional Radiology" held at the British Institute of Radiology on 28 March 2007 as a template to discuss recent progress in this area, some current problems and plans for the future.

Commentary

Radiation risks in interventional radiology

1A Coppice Avenue, Great Shelford, Cambridge CB22 5AQ, UK

Correspondence: P P Dendy, 1A Coppice Avenue, Great Shelford, Cambridge CB22 5AQ, UK. E-mail: dendy{at}waitrose.com

The number, diversity and complexity of interventional radiological examinations have all increased markedly in recent years, and it is widely recognized that some of these procedures carry greater risks than many other radiological procedures. This Commentary uses a meeting on "Radiation Protection in Interventional Radiology" held at the British Institute of Radiology on 28 March 2007 as a template to discuss recent progress in this area, some current problems and plans for the future.

	Introduction

Top Abstract Introduction Current trends in interventional... Doses to patients Staff doses Equipment factors and programme... Education and training Conclusions and future work References

 
It is widely recognized that some procedures in interventional radiology (IR) carry greater radiation risks than many other radiological examinations. There are increased stochastic risks to the patient from long screening procedures and a greater risk of deterministic effects, generally to the skin, especially when the radiation beam is directed at the same area on the skin surface for most, if not all, of the examination. There are also increased stochastic risks, and possibly deterministic risks, to staff who are close to the patient from scattered radiation. For all of these reasons, the European Council Directive of 30 June 1997 on "Health protection of individuals against the dangers of ionising radiation to medical exposure, and repealing Directive 84/466 Euratom" [1] identified IR as a "Special Practice" requiring special attention to be given to quality assurance programmes, quality control measures and patient dose assessment.

Several publications give information and/or guidance on radiological protection in IR (see [2–4]). However, in the past 4 years, the number, diversity and complexity of IR examinations have all increased markedly. Therefore, it is timely to review recent progress, current problems and plans for the future. This Commentary is based on a meeting held at the British Institute of Radiology on 28 March 2007 to address these issues. The multidisciplinary nature of many of them was emphasized.

	Current trends in interventional radiology — the radiologists' viewpoint

Top Abstract Introduction Current trends in interventional... Doses to patients Staff doses Equipment factors and programme... Education and training Conclusions and future work References

 
To put the subject into a clinical context, it is helpful to review the state of the art in IR. The first interventional procedure is generally attributed to Dotter in 1964 [5] who used dilators to relieve arterial stenosis. Since that time, proliferation of IR procedures has proceeded apace, often driven by technical advances. By 2000, the compound annual rate of increase in IR across all service settings in the UK was 10%.

Some recent trends in IR were identified by Hamish Ireland (Edinburgh):

• A major reduction in angiograms — at Edinburgh Royal Infirmary, the number of femoral arteriograms fell from 566 cases in 1999 to 42 in 2006. This demise results primarily from 3D imaging techniques, notably MR angiography, duplex ultrasound and CT angiography. Space in angiographic rooms has been filled by interventional procedures, which are generally more complex, take longer to perform and involve more radiation
  
• Multimodality guidance – e.g. the combined use of ultrasound and fluoroscopy for venous access, abscess and nephrostomy. The technique is precise, real time, and uses less ionizing radiation.
  
• Pre-intervention — other modalities are being used for the diagnostic aspect of the interventional procedure, e.g. CT angiography to localize gastrointestinal haemorrhage prior to embolisation.
  
• Increasing jugular/neck/arm access
  
• Increasing liver access — biliary stents and portal vein embolisation.
  
• More IR procedures performed by non-radiologists — e.g. cardiologists and surgeons.
  

Ireland also indicated some drivers for change in the use of IR procedures:

• Changes in incidence of disease — the ageing population and increasing obesity are both leading to increased vascular disease and cancers, resulting in more peripheral angioplasty and stents, more biliary ureteric stents and more radiofrequency (RF) ablation. Increasing alcohol intake and hepatitis C have increased cirrhosis of the liver, resulting in more chemoembolisation processes and transjugular intrahepatic portosystemic shunts (TIPS).
  
• Perceived effectiveness — new procedures and drugs are often received with great enthusiasm and their benefits overstated. This is followed by a period of realization that the efficacy is not as high as claimed. Hopefully, the "effectiveness" stabilizes at a lower level, from which it can improve steadily. Many IR procedures are in the first phase of this cycle. This raises the problem that it is very difficult to get sufficient patient numbers for rigid verification of the benefit of a new technique, and "justification" within the framework of ionising radiation regulations may need re-appraisal.
  
• Fashion — this is related to the enthusiasm noted above.
  
• New technology — better imaging, e.g. ultrasound and CT guidance to aid targeting, and better procedure kits for endovascular aneurism repair (e.g. RF ablation) will be developed. Gradual introduction of robotic systems can be expected, especially in view of their potential for savings on staff doses if the operator can be further from the patient.
  

Using endovascular techniques as a specific example of IR procedures, Anthony Nicholson (Leeds) expanded on some important issues in radiation protection for the radiologist. IR plays a major role in accident/trauma and is becoming an increasing part of emergency work [6, 7]. In some situations, the technique is far more effective than surgery for stopping massive bleeding. Significant doses of radiation are involved, including top-to-toe CT scanning beforehand, a flush aortogram to find the exact site of bleeding and CT scans to check that all is well subsequently, in addition to the radiation required for the interventional procedure itself. Is all this ionizing radiation necessary? Nicholson said that in some cases it was, and suggested that a simple working definition of optimized dose in IR was "the least amount of radiation to get the job done". However, interventionalists must not get carried away by their enthusiasm to utilize fully this new technology now available to them. For example, it may be feasible to follow up a TIPS using ultrasound, with no further need for CT digital subtraction angiography.

For the radiologist, there is uncertainty and some confusion over the most suitable measure of doses to patients in IR. Two points need to be kept in mind. Firstly, the measure required to assess stochastic risk will normally be different from the measure best-suited to assess deterministic risk. Secondly, the pattern of exposure on the skin surface may be entirely localized to one area, e.g. in neuro work, or spread over a larger area, e.g. cardiology. For comparison of performance, cumulative dose (CD) (discussed in the next section) is often preferable to dose-area-product (DAP), and Nicholson expressed concern at the wide range of CDs reported in the literature. For example, as part of a large multi-centre study, Miller et al [8] recorded CDs from 135 TIPS cases. The range was 0.1–8 Gy, with 26% <1 Gy but 21% >3 Gy. This wide range in recorded doses has been reported many times for several different procedures. Setting aside equipment-related factors, which are discussed later, several contributing factors can be identified: patient factors (including complexity of the procedure), poor or no training of the interventionalist, experience of the interventionalist, high/low volume case load, and poor technique. However, in the final analysis, it is difficult to tease out the dominant causes of variation from this list.

	Doses to patients

Top Abstract Introduction Current trends in interventional... Doses to patients Staff doses Equipment factors and programme... Education and training Conclusions and future work References

 
Elisio Vañó (Madrid) gave an overview of doses to patients. After reviewing the relevant recommendations (e.g. [2–4]), he made some important points.

Firstly, DAP, which is a measure of the total energy imparted to the patient as ionising radiation, correlates reasonably well with the stochastic risks of induced cancer probability or hereditary effects. Variations in total energy absorbed with kVp, filtration and tissue weighting factors (w_T) have a relatively minor effect. Conversely, for many interventional procedures, DAP is a poor measure of the risk of deterministic effects, especially for cardiology where there tends to be several non-overlapping fields. Therefore, it is important to record the CD or cumulative air KERMA (kinetic energy released per unit mass) at the interventional reference point (IRP). Note that manufacturers' data will normally be given in terms of air KERMA dose rate under specified output conditions in a scatter-free measurement at the IRP, chosen to approximate the skin surface. For a system with an isocentre, the IRP is a point on the reference axis 15 cm from the isocentre in the direction of the focal spot [9].

Secondly, the beneficial uses of diagnostic reference levels in plain film radiology are now well established:

• to compare the practice, in terms of level of radiological risk, with that in other centres
  
• to decide if there is a margin for optimization in terms of the X-ray system settings or protocols
  
• to detect abnormal situations with high radiological risk for the patient.
  

Can the principle of reference levels (RLs) be transferred to interventional procedures and would this be an aid to optimization in IR? The starting point is to collect detailed information on a wide range of procedures from a number of different centres. Some preliminary data, using equipment supplied by only one manufacturer, are shown in Table 1 for both DAP and CD.

Further information on patient skin dose monitoring was given by Rachel Morrell (Nottingham). A review of the advantages and limitations of the various methods available shows that none is ideal:

• DAP is measured with an ionization chamber that covers the whole beam area. It is easy to use and measures the total energy deposited. However, it may not indicate localized skin dose.
  
• Fluoroscopy time has the merit of convenience but gives no information on dose distribution and is not a good indicator of dose, as the dose from screening is usually small compared with the dose from digital image acquisition.
  
• A skin dose indicator gives the cumulative dose to a fixed point in the X-ray beam, so it is the highest possible value of the dose to a fixed point on the skin. Again, it is convenient to use but gives no information about dose distribution.
  
• Thermoluminescent dosemeters (TLDs) are accurate, linear over a wide range of doses and re-usable, but may miss the peak dose (depending on position). They are also easily damaged and time-consuming to use.
  
• Scintillation monitors and diodes have the advantages of real-time read-out but have poor area coverage and are likely to be visible on the image.
  

International Commission on Radiological Protection (ICRP) Report 85 [2] recommends that the magnitude and position of the maximum skin dose should be recorded in the patient's notes if it exceeds 1 Gy for procedures that are likely to be repeated, or 3 Gy for all procedures. Patients receiving more than 1 Gy to the same skin area can be considered to be "at risk" from deterministic effects and should be informed about potential symptoms and appropriate action they should take if any skin changes occur.

Morrell and Rogers [11] have made a detailed study of the use of Kodak EDR 2 film (Eastman Kodak Company, Rochester, NY) for skin dose monitoring. Measurements were made for 20 coronary angiography and 32 PTCA procedures. The film gives good information on the pattern of irradiated fields across the patient's back, with variations in radiation blackening. For coronary angiography, all skin doses were well below 1 Gy. However, 23% of PTCA patients received skin doses of 1 Gy (the dose at which the film saturated) or more. The authors concluded that practical compliance with ICRP recommendations requires a robust method for skin dosimetry that is more accurate than DAP and applicable over a wider range of doses than EDR2 film. Gaf chromic films (International Specialty Products, Wayne, NJ) have higher dose ranges than EDR2 film and do not require processing, but at present they are prohibitively expensive for routine dosimetric use.

There are indications that the problem of comprehensive and relevant dosimetry may be amenable to solution by mathematical modelling. For example, a mathematical model developed by Morrell and Rogers [12] uses exposure and projection data from the acquisition runs stored in the DICOM image files to estimate skin doses. Early work showed that maximum skin doses calculated by this model correlated well with those measured on EDR2 film. Three methods for including the relatively smaller dose contribution from fluoroscopy were investigated and all successfully identified patients receiving skin doses in excess of 1 Gy. Two advantages over film dosimetry are that doses above the saturation point of film can be assessed and there is less staff involvement. Furthermore, in theory, the method can be combined both with data on doses to organs and w_T values to give an estimate of effective dose to patients, and with data on room scattering to provide information on staff doses. Some of the technology that would be required for a comprehensive dosimetric system is not yet available, but this might be a fruitful area for discussion with manufacturers.

	Staff doses

Top Abstract Introduction Current trends in interventional... Doses to patients Staff doses Equipment factors and programme... Education and training Conclusions and future work References

 
Vañó emphasised that "appropriate" staff dosimetry is essential, but what is "appropriate"? There are several possibilities:

• one personal dosemeter (usually under the apron)
  
• two personal dosemeters (one under, one over the apron)
  
• additional dosemeters for the extremities
  
• electronic dosimetry
  
• research occupational dosimetry (with several TLDs)
  

There are also practical problems with occupational dosimetry:

• staff forget to wear their dosemeters
  
• there are mistakes in the use of dosimeters above and below the apron — perhaps they are interchanged?
  
• dosemeters are worn that were assigned to other interventionalists
  
• there may be uncertainty over whether protective devices have been in place or worn, e.g. lead curtains, thyroid shields, protective goggles.
  

Nicholson commented that, in the UK, the Regulations did not make it easy for staff to monitor themselves and cited the problem of "Radiation Passports" for interventionalists who work at more than one location. There is general agreement that the interventionalist will be giving 100% of their attention to the patient with little thought for the radiation dose to themselves. However, this should not prevent the wearing of dosemeters as directed.

Mark Whitby (Glasgow) gave further information on staff dose and monitoring requirements. Many factors will influence the amount of scattered radiation reaching staff, including the size of the patient, the equipment and its set-up, the type of procedure (especially its complexity and length), tube geometry, position of staff (especially their hands) relative to the patient, the skill and experience of the operator and the effective use of radiation protection principles and devices.

Given the uncertainties, it is not surprising that Whitby's data showing indicative doses for IR/cardiology covered wide ranges (Table 2). It is clear that, at the upper ends of these ranges, staff will need to be "Classified Radiation Workers" and some form of monitoring is essential. The challenge is to ensure that it is both relevant and proportionate to the risk. For whole body monitoring, there is still uncertainty as to whether two monitors (one below the apron at, for example, chest level and one above the apron at collar/eye level) are necessary. Whitby suggested that one monitor should suffice for a low work load, with two monitors for a higher work load. The boundary between low and high workloads should be set in the region of 1–5 mSv per annum for body doses. The second monitor will also be necessary if eye doses are high, i.e. with a boundary in the range 10–50 mSv per annum. Unfortunately, it is difficult to specify the boundary in terms of readily available machine output data such as DAP, because the precise position of the operator is also a factor. Therefore, a conservative approach to monitoring must be taken initially. This can often be relaxed when real monitoring data become available.

Because of the uncertainties involved, the hands should always be monitored when a procedure in which the hands have to be close to the main beam is used for the first time. Subsequently, monitoring might be relaxed to "periodic" if the estimated annual hand dose is less than 50 mSv per annum, or stopped altogether if below 10 mSv per annum. For further information on recommended extremity monitoring at different dose levels, see Martin and Whitby [13].

Proper use of protective measures can have a big impact on staff doses and is essential e.g. see Vañó et al [14]. In the example cited, nine staff radiologists and eight interventional cardiology fellows working in a centre performing 5000 procedures a year were monitored using personal dosemeters — below and outside lead aprons — over a period of 15 years. At the start of the study, maximum doses of 100–300 mSv per month above the apron and 5–11 mSv per month (mean 10.2 mSv per year) below the apron were recorded. In the past 5 years, after a programme of optimization, mean occupational doses under the lead apron have fallen to 1.2 mSv per year (14% of those in 1989 to 1992), whereas doses recorded outside the lead apron have fallen by a factor of 14. The most effective actions for reducing the radiation risk were (i) training in radiation protection, (ii) a programme of patient dose reduction, and (iii) the systematic use of radiation protection facilities, specifically ceiling-suspended protective screens.

	Equipment factors and programme set-up

Top Abstract Introduction Current trends in interventional... Doses to patients Staff doses Equipment factors and programme... Education and training Conclusions and future work References

 
Taking the title "Designing radiation protection into the system", Nick Marshall (London) gave an excellent review of the importance of well-designed equipment in dose reduction. Features of the equipment that are known to affect dose include:

• Spectral filtration — 0.1 mm Cu can reduce skin dose rate by 40% (but has less impact on effective dose), increases tube loading and reduces contrast. A range of filters should be available for different programmes, and the spectral filtration should be displayed to help with trouble-shooting and optimization.
  
• Pulsed fluoroscopy — is virtually essential for interventional work. In modern systems, the high tension is kept constant and a third grid controls the flow of electrons to the anode. Optimization of fluoroscopy pulse widths and careful choice of entrance exposure/pulse during calibration of equipment is important. Hernanz-Schulman [15] reported that, in paediatrics, grid-controlled pulsed fluoroscopy at 3.75 pulses s^–1 can reduce dose by ten-fold without significant reduction of contrast or spatial resolution.
  
• Organ curves — the X-ray set can respond to demand for increased output by increasing kV, mA or a combination of both. A range of options will be available so it is important to know the impact of each on contrast and dose.
  
• Image receptor — there is increasing emphasis on solid-state detectors (flat panel detectors (FPDs)), using either direct conversion with amorphous selenium or indirect conversion using caesium iodide, in preference to image intensifiers. The FPDs have a better dynamic range because there is no television camera. Also, the manufacturers claim a better detective quantum efficiency, which could be used either for dose reduction or to give better signal-to-noise ratio. However, Marshall displayed data showing little difference in input dose rate at the image receptor under automatic exposure control for image intensifier and FPD systems. He suggested that the failure to demonstrate any advantage might be due to electronic noise in the FPDs.
  

Marshall also made some important general remarks on radiation protection implications of the interface between radiology staff and manufacturers:

• Before entering the market, the radiology department should specify, as fully as possible, the scope of interventional procedures the equipment will be performing. A system can only be "well-designed, reliable, flexible, everyone likes using it" if it is fit for the clinical purpose for which it is intended. Manufacturers often have very little data on dose but the "Safe Specs" system provided by KCARE (King's Centre for the Assessment of Radiological Equipment) contains a useful questionnaire to aid evaluation. "Homework" at this stage is likely to have a pay-back in terms of exposure times and dose rates.
  
• The architecture of the programme set-up system must be fully understood. For example, is it a free-standing system in which the fluoroscopy mode is completely separate from the examination type or does the choice of examination pre-define the acquisition and fluoroscopy parameters? In either case, a wide range of operational modes and examination pre-sets (input dose rates, pulsed fluoroscopy setting, spectral filtration) must be correctly set up to ensure adequate image quality and doses as low as possible for all examinations.
  
• For pre-defined studies, proper liaison must be maintained between the radiology department, the applications specialist employed by the manufacturer and the medical physicists involved in commissioning. The commissioning physics tests are often based on factory default settings and are performed before the applications specialist adjusts the software for specific clinical applications. Further commissioning should be done, for example, 6 months later to establish actual clinical baselines.
  

Careful attention to detail in setting parameters and clinical observations can aid in optimization and indicate further areas of study. These principles were well illustrated in two case studies in interventional neuroradiology presented by Halina Szutowicz (Cambridge).

A Siemens Axiom Artis BA was installed in Cambridge in May 2004, and each of the factory settings was challenged with alternative options. For example, 15 pulses s^–1 was reduced to 7 pulses s^–1 but gave jerky movements and was rejected. Dose rates per pulse were reduced from 23 µGy, 32 µGy and 45 µGy per pulse to 15 µGy, 18 µGy and 23 µGy per pulse, respectively. Arrangements were made to put the dose record file with the stored images, and a total dose record was placed in the patient's radiology record.

The first case was a long procedure, lasting a total of 9 h, to embolise a single large occipital arterio/venous malformation. 2.5 weeks later, the patient reported sudden loss of hair from the back of the head, but no apparent erythema at this stage. Subsequently, marked erythema and almost complete epilation developed.

A number of actions were taken:

• The entrance field from the postero-anterior (PA) plane was accurately matched
  
• Dose readings were retrieved — the PA entrance skin dose (ESD) was 8.1 Gy
  
• The radiation protection advisor and radiotherapy colleagues were contacted
  
• ESD was checked against published threshold doses for deterministic effects
  
• Embolisation programmes were re-programmed to lower dose rates
  
• Radiation protection survey reports were checked
  
• DAP meters were recalibrated to a tighter tolerance band (5%)
  
• Post-procedure patient visits were introduced when the estimated skin dose exceeded 2 Gy.
  

3 months later, the patient's hair had regrown. This is consistent with temporary epilation. Perhaps there was early erythema that went unnoticed?

Fewer details were given on the second case, except that it was a spinal embolisation. The lateral plane ESD was off scale (maximum 999 mGy) and had to be estimated from the DAP reading at 11.3 Gy. At 10 months, telangiectasia was visible, with perfect collimation to the irradiated site, but the patient had not noticed any of the early effects he had been told to expect.

The work raises some interesting questions for future study:

• How accurate are deterministic time scales? Much of the available data comes from radiotherapy with fractionated exposures
  
• In both cases, why were minor effects not more apparent earlier?
  
• Can anything be done in advance to minimize deterministic effects in such patients, e.g. vitamin C as an antioxidant?
  

	Education and training

Top Abstract Introduction Current trends in interventional... Doses to patients Staff doses Equipment factors and programme... Education and training Conclusions and future work References

 
Several other matters that deserve mention will be grouped together under this heading.

As Claire Cousins (Cambridge) pointed out, very few radiologists or interventionalists will see a radiation injury during their working lifetime and will never know if their use of radiation has caused cancer. However, the fact that "these things happen to someone else" is not an acceptable reason for ignoring radiation doses.

Long-term follow-up of atomic bomb survivors has shown that an excess risk of cancer is associated with a few tens of mSv [16]. At the much lower doses associated with many diagnostic procedures, e.g. those comparable with annual background radiation, extrapolation is more uncertain, not least because homeostatic mechanisms may reduce the radiation effect [17]. However, at the level of effective dose (E) typical of IR, the analysis of Brenner and Elliston [18] is pertinent. They assumed an E value of 12 mSv for a single full whole-body CT examination (lung organ dose 15 mGy) and estimated a cancer risk of 8x10^–4 for a 45-year-old, which falls slightly with age. Adequate information must be given to all relevant staff on such levels of radiation risk, which are typical of many IR procedures, but it is important to emphasize that there are very large sources of error in using the effective dose concept for risk estimation [19]. The 95% confidence interval on the Brenner and Elliston risk estimation [18] was about a factor of three each way, and so numerical estimates of excess risk of fatal cancer derived from E should not be included in patient reports.

A further important aspect of training is to put the radiation risk into perspective with other everyday risks. For example, the risk of death in a car accident in the US in 1999 was about a quarter that of the hypothetical CT scan discussed above, but clearly a series of whole-body CT scans, or comparable IR studies, will accrue a significant risk.

If risk estimation in IR is difficult, then estimating the benefit is infinitely more difficult. Clearly, there are often massive benefits for the patient but we have not even begun to quantify the level at which benefits and radiation risk might be of a similar magnitude. In the absence of a sound methodology, it is difficult to know what to teach. Because, in general, the radiation is being used as an aide to therapy in IR, there is perhaps something to learn from the approach adopted in radiotherapy.

Vañó discussed the work that has been done in Europe on the theoretical training required [20]. A syllabus has been prepared and a recommended teaching time allocated to each subject for different staff groups, e.g. diagnostic radiology specialists, IR specialists, other medical doctors using X-ray systems, radiographers and maintenance engineers. Total teaching varies from 10 h to 60 h. The International Atomic Energy Agengy (IAEA) has produced specific training material for cardiologists and held regional courses in several continents. An ICRP Working Party is currently preparing definitive recommendations with a tentative title "Radiation protection training for physicians that conduct or order diagnostic and interventional procedures using ionising radiation".

On the practical side, Nicholson expressed concern that too many doctors are becoming involved in IR without proper structured training. Ensuring low doses to patients and staff should be an integral part of good technique, and it is a little surprising that there was no mention of radiation dose or radiation risk in a 2002 publication on IR education in Seminars in Interventional Radiology [21]. The article on "Virtual Reality Training in Interventional Radiology" in that issue would have been a good opportunity to discuss how the technical variables that affect patient and staff doses might be factored into such training.

Another problem is accreditation. It may be feasible for training centres to be accredited by National Professional Bodies but how will the "students" be assessed and how does one decide the level at which they can be considered competent? Multiple choice questions seem entirely inadequate.

There is also much to learn for the radiographer in the IR room. Kevin Walker (Stirling) suggested that because of their background and previous training, the radiographer is best suited to be the Radiation Protection Supervisor (RPS). This could be a good arrangement provided that the radiographer is experienced and has a "hands-on" role in the interventional room on a regular basis. He or she must not only be responsible for their own training, i.e. the equipment, quality assurance, optimization of procedures and importance of interdisciplinary work, but must also supervise the training of others. New staff need to know about stochastic and deterministic effects, scattered radiation, protective devices, how to wear their dosemeters and dose limits. The RPS must also ensure that other staff, who are preoccupied with the procedure and the patient, do not inadvertently and unnecessarily increase their own doses, and must be generally pro-active in radiation protection matters. With the future developments in "skill mix", other individuals may be better equipped to be the RPS in certain situations.

	Conclusions and future work

Top Abstract Introduction Current trends in interventional... Doses to patients Staff doses Equipment factors and programme... Education and training Conclusions and future work References

 
To sum up briefly:

• Demographic changes and changes in patterns of disease will ensure that the number of IR procedures, and sometimes their complexity, will continue to increase. Occasionally, this will result in higher doses to individual patients. It will certainly increase the radiation burden to the population as a whole.
  
• Better imaging equipment, procedure kits and dose-saving devices will be introduced by the manufacturers, perhaps with an emphasis on robotics.
  
• Work on patient doses will continue, not least because this is a legal requirement. Manufacturers should ensure that digital technology is exploited for management of the data.
  
• ICRP will be widening the use of reference levels and the British Institute of Radiology will be setting up a Working Party to look at skin doses. DAP and CD are useful quantities when comparing doses for similar procedures.
  
• Careful follow-up of patients undergoing lengthy procedures may provide important new information on deterministic effects.
  
• Greater efforts should be made to relieve the operator of the responsibility for their own monitoring. Perhaps, as a starting point, the mathematical modelling described by Morrell and Rogers could be used to produce maps of scattered radiation.
  
• The concept of effective dose should be used with caution for estimating radiation risk because of the large errors and uncertainties involved.
  
• Training and education will continue to receive the highest priority. ICRP will finalize its recommendations for training doctors and the IAEA will continue to promote programmes in staff training, but some key issues need to be addressed: (i) how does one quantify the clinical outcome, i.e. the "benefit" in IR, and (ii) how should the practical competence of new operators be verified?
  

Received for publication June 11, 2007. Revision received August 30, 2007. Accepted for publication August 31, 2007.

	References

Top Abstract Introduction Current trends in interventional... Doses to patients Staff doses Equipment factors and programme... Education and training Conclusions and future work References

 

1. European Union, Council Directive 97/43/EURATOM of 30 June 1997. Health protection of individuals against the dangers of ionising radiation in relation to medical exposure (repealing Directive 84/466/EURATOM, O.J. No L 265/1, 5.10.1984). Luxembourg: Official Journal of the European Communities 1997. L180:22–27.
2. International Commission on Radiological Protection, Report 85. Avoidance of radiation injuries from medical interventional procedures. Ann ICRP 2000; 30.
3. IAEA. Radiological protection of patients in diagnostic and interventional radiology, nuclear medicine and radiotherapy. Proceedings of an International Conference; 2001 March; Malaga, Spain. Vienna: IAEA.
4. IPEM Medical and Dental Guidance Notes. A good practice guide on all aspects of ionising radiation protection in the clinical environment. York, UK: Institute of Physical Sciences in Medicine; 2002.
5. Dotter CT, Judkins MP Transluminal treatment of arteriosclerotic obstruction. Description of a new techic and a preliminary report of its application. Circulation 1964;30:654–70.[Abstract/Free Full Text]
6. Surash S, Robertson I, Calvey AJ, Kessel D, Patel J, Kent PJ, et al. An audit of out of hours interventional vascular radiology. Eur J Vasc Endovasc Surg 2003;25:573–7.[CrossRef][Medline]
7. Funaki B. On call treatment of acute gastrointestinal haemorrhage. In Van Ha TG, editor. Interventional Radiology on Call. Sems in Interven Radiol 2006;23:215–22.[CrossRef]
8. Miller D L, Balter S, Cole P E, Hollington T L, Schueler B A, Geisinger M, et al. Radiation doses in interventional procedures. The RAD-IR Study Part 1: overall measures of dose. J Vasc Interv Radiol 2003;14:711–28.[Medline]
9. British Standard: medical electrical equipment. Part 2–43: particular requirements for the safety of X-ray equipment for interventional procedures. BS EN 60601-2-43; 2001.
10. Bernardi G, Padovani R, Morocutti G, Vañó E, Malisan MR, Rinuncini M, et al. Clinical and technical determinants of the complexity of percutaneous transluminal coronary angioplasty procedures: analysis in relation to radiation exposure parameters. Catheterisation and Cardiovascular Intervention 2000;51:1–9.[CrossRef]
11. Morrell RE, Rogers AT. Kodak EDR2 film for patient skin dose assessment in cardiac catheterisation procedures. Br J Radiol 2006;79:603–7.[Abstract/Free Full Text]
12. Morrell RE,Rogers AT. A mathematical model for patient skin dose assessment in cardiac catheterisation procedures. Br J Radiol 2006;79:756–61.[Abstract/Free Full Text]
13. Martin CJ, Whitby M. Application of ALARP to extremity doses for hospital workers. J Radiol Prot 2003;23:405–21.[CrossRef][Medline]
14. Vañó E, Gonzales L, Fernandez JM, Alfonso F, Macaya C. Operational radiation doses in interventional cardiology: a 15 year follow up. Br J Radiol 2006;79:383–8.[Abstract/Free Full Text]
15. Hernanz-Schulman M. Can fluoroscopic radiation dose be substantially reduced? Radiology 2006;238:1–2.[Abstract/Free Full Text]
16. Preston DL Shimizu Y, Pierce DA, Suyama A, Mabuchi K. Studies of mortality of atomic bomb survivors. Report 13. Solid cancer and non-cancer disease mortality. Rad Res 2003;160:381–407.[CrossRef][Medline]
17. Dendy P, Brugmans MPJ. Commentary: low dose radiation risk. Br J Radiol 2003;76:674–7.[Free Full Text]
18. Brenner, Elliston CD. Estimated radiation risks potentially associated with full body CT screening. Radiology 2004;232:735–8.[Abstract/Free Full Text]
19. Martin CJ. Effective dose: how should it be applied to medical exposures? Br J Radiol 2007;80:639–47.[Abstract/Free Full Text]
20. Radiation protection 116. Guidelines on education and training in radiation protection for medical exposures, 2000. Available from: http://europa.eu.int/comm/energy/nuclear/radioprotection/publication_en.htm [Accessed 5 November 2007]
21. Interventional Radiology Education. In Kaufman JA, Rösch J, editors. Seminars in Interventional Radiology; 2002; 19.

This Article Abstract Full Text (PDF) Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Dendy, P P Search for Related Content PubMed PubMed Citation Articles by Dendy, P P