British Journal of Radiology (2006) 79, 326-330
© 2006 British Institute of Radiology
doi: 10.1259/bjr/36530782
A review of current local dose–area product levels for paediatric fluoroscopy in a tertiary referral centre compared with national standards. Why are they so different?
M P Hiorns,
A Saini and
P J Marsden
Radiology, Great Ormond Street Hospital, Great Ormond Street, London WC1N 3JH, UK
 |
Abstract
|
|---|
A prospective single centre study has been performed to assess dose–area product (DAP) values in children having fluoroscopic examinations and to revise local diagnostic reference levels (DRLs). DAP measurements for 2658 examinations performed in a dedicated fluoroscopy room over a period of 21 months were analysed. Data for the eight most commonly performed examinations (2215 cases) are presented. DAPs (75th centile) for upper gastrointestinal studies and micturating cystograms are substantially lower (by a factor of between x 5 and x 25) than the current national reference doses (NRDs), with some of the median values being 50 times lower. The small DAP values in all examinations demonstrate the substantial reduction in dose and consequent risk that can be achieved when both equipment performance and operator technique are optimized. Whilst we recognize that different institutions will have differing practices, it is important that practitioners are aware of the range of DAPs achievable and that NRDs do not necessarily represent best practice, and may falsely reassure.
|
Introduction
|
|---|
The risks of ionizing radiation are higher in the paediatric population than in adults [1, 2]. It is therefore particularly important that the dose–area product (DAP) used in imaging children should be as low as practicable whilst providing the clinician with diagnostic information [3]. The actual doses achieved will vary greatly depending on the age, sex, body mass, body thickness and cooperation of the child. The type of equipment and its use by the operator will also affect dose levels. The most recent national guidelines published by the National Radiological Protection Board (NRPB) in 2002 which include paediatric procedures for the first time are based on a review of dose measurements collected over a 5 year period (1996–2000). DAPs were measured from 3671 paediatric fluoroscopic examinations in up to 29 different rooms [4]. From the results, recommended NRDs could only be defined for three paediatric fluoroscopic examinations; micturating cystograms, barium swallows and barium meals. To our knowledge, very little published data exists on the current range of doses being delivered throughout the country.
The objectives of this study were to revise locally established DRLs and to compare performance with the current national guidelines.
|
Methods
|
|---|
Data were collected prospectively on consecutive patients from September 2001 to May 2003. For each examination performed the patients name, hospital number, date, kV, mAs, screening time, DAP, examination type, and radiologist's name were recorded. All studies were performed on a Siemens Polystar digital unit (installed 1999; Siemens, Erlangen, Germany). The DAPs were recorded in cGycm2 using a PTW Diamentor M4 DAP meter, which is specifically sensitive enough for paediatric work (PTW, Freiberg, Germany) with a high resolution of 0.01 cGycm2. The DAP meter is calibrated by our radiation physicists on a yearly basis and a tolerance of ±3% is considered acceptable. Overall uncertainty is within ±25% as recommended in the National Protocol [5]. Other routine quality assurance is completed every 3 months.
The exact technique was at the discretion of the radiologist. For the period surveyed, a total of nine consultant radiologists and several specialist registrars, of varying experience, performed the examinations. Views obtained were tailored to the individual clinical question but basic views were obtained in all patients, these would include a mixture of grabbed images and spot exposures. Typically, both an upper gastrointestinal (GI) series and a micturating cystogram would comprise between six and 10 images.
|
Results
|
|---|
Data were collected in 2658 cases on an individual basis, but were then grouped by age for specific examinations. Only one study were abandoned, during the period and all other studies produced images of diagnostic quality. All the data were analysed, but the data for the eight most common examinations are presented (2215 cases). As patient dose has been shown to increase with body size [6], patients were assigned into one of three groups; 0–12 months, 1–7 years inclusive and 8 years plus. The case distribution for our institution is shown (Figure 1
) and data for the most common examinations are listed (Table 1). The DAP range, mean DAP, standard deviation, median DAP, 75th percentile and 90th percentile were all calculated. These values were used to establish revised DRLs for our institution (Table 2). The 75th percentile was used for the DRL (bold type), but we have also added median values in our local reference chart as this gives a good indication of overall practice and is less skewed by outliers. Radiologists are quickly able to see if their DAPs are similar to those of their colleagues. Some of our examination groups contained only small numbers, but as a large referral centre this is a reflection that very few of these examinations are done in any centre. In these cases we include the figures as a guide only.
View larger version (35K):
[in this window]
[in a new window]
|
Figure 1. Case distribution of examinations in our institution. GI, gastrointestinal; MCU, micturating cystourethrography; FT, follow through; IVU, intravenous urogram; NJT, nasojejunal tube.
|
|
View this table:
[in this window]
[in a new window]
|
Table 2. Revised DRLs used in our institution. The 75th percentile values are demonstrated in bold. The values below represent the median dose–area product (DAP) for each examination. Children over the age of 8 years having contrast or barium enemas have been included together due to the low numbers in each group
|
|
|
Discussion
|
|---|
In the UK, the adoption of the European Council Directive 97/43 Euratom [7] through the Ionising Radiation (Medical Exposure) Regulations [3] has made the establishment of DRLs and their update through clinical audit a mandatory requirement. National diagnostic reference levels have been established for a variety of mainly adult examinations as a guide to what is thought to represent "good and normal practice". Individual hospitals can then compare their DAP performance and modify their practice accordingly. The low numbers of patients and their distribution over the range of different ages and sizes limits the number of paediatric DRLs currently available. Nevertheless, it is important that hospitals in which a relatively low number of studies are performed continue to assess local DAP levels and have some guide to those used elsewhere.
The NRDs as recommended by the NRPB are based on the 75th percentile of their survey data and are shown in Table 3 [4]. The NRPB stratified its data into five standard sizes of children corresponding to newborn babies, 1 year, 5 years, 10 years and 15 years old and children were allocated according to the nearest milestone. Dose levels, either in the form of entrance surface doses (ESD) per radiograph or DAP per examination were adjusted for body thickness or both the height and weight to give a value (in cGycm2) corresponding to the nearest standard size child. It can be seen that for all types of examination there are only small differences between the DAPs for the 1 year and 5 year old groups. The NRPB has therefore recommended that the same NRD be applied for both age groups. The establishment of reference doses in paediatric radiology as a function of patient size has been investigated for patients of the same age groups as used by the NRPB and it has been shown for plain film radiography in the trunk that, although there may be a significant difference in size between adjacent reference ages, normalization factors for any intermediate size were unlikely to exceed a factor of two [8, 9].
View this table:
[in this window]
[in a new window]
|
Table 3. Recommended national reference doses(NRDs) for complete examinations on paediatric patients based on the NRPB 2000 review. Recommended values for 1 year and 5 year old sizes are given in brackets. Comparisons are given with the nearest local equivalent dose reference levels (DRLs)
|
|
The practical disadvantage of applying a wide variety of standard sizes for dose assessment is more evident in non-specialist hospitals where patient numbers are small and long periods of data collection would be necessary. Since body size affects the dose received in any radiological procedure it would seem ideal to assign patients into narrow age groups that will not have an unduly large range of body weight and size [10]. Whilst we realise that the use of wider age groups may result in a larger range of doses, and obscure the effects of examination technique and equipment performance, the age groups used in this study have been designed for simplicity and for consistency with previous publications. Despite the use of a single dedicated fluoroscopy suite, there is a wide variation in measured DAPs within the individual age groups of all the examinations in our study. Such variation is similar to that found by both the NRPB and the Council of the European Communities (CEC), but in these studies the differences were largely attributed to variations in equipment and film–screen combinations [4, 11]. In our study, the range reflects the wider distribution of patient size, the technical difficulties of imaging uncooperative children and the variation of operator technique and experience (including training specialist registrars). We have found that some operators are consistently producing diagnostic quality studies at DAP levels significantly lower than others.
In our institution barium meals and barium swallows are combined as one procedure, the upper GI series. We have compared our values with those given for both barium meals and barium swallows (Table 3
) and have found that our closest equivalent DAPs for 75th percentile are between 9 and 25 times lower (UGI in patients 15 years, and <12 months, respectively) and for a micturating cystogram are 5 to 18 times lower (MCU in patients of 8 years and older, and <12 months, respectively). Reviewing our range of DAPs shows that, for example, in UGIs in children up to 1 year, even our 90th centile value (14.8 cGycm2) is 13 times less that the NRPB 75th centile (with the total range being 0.06–41.9 cGycm2).
The mean DAP results from the present study are also substantially less than similar data from previous studies, including another specialist centre [12, 13] (Table 4). A further study published in 2000 [8], collecting DAP data from 12 European hospitals, investigated doses for micturating cystograms in patients between neonate and 15 years and obtained third quartile doses approximately 12 times higher than the figures we have obtained for all age groups. A study in Finland [14] considered a total of 217 patients having a wide range (12 types) of fluoroscopic examinations and derived third quartile DAP values for three of these examination types, which again are approximately 12 times higher than our figures for comparable examinations and age groups. However, it is important to note that these studies were conducted in the 1990s using conventional fluoroscopic equipment. The large differences in doses demonstrated are almost certainly due to differences in equipment and operator technique.
The use of automatic post processing of data in digital systems prior to image display does not provide the operator with feedback on exposure levels and as such has the potential to increase dose. In our experience this is generally not the case and we have found that the use of low frame rates and last image hold ability is advantageous. A similar trend was observed by the NRPB who found that in 8 of 16 types of examination in adults, the use of digital spot imaging reduced mean DAP values by between 20% and 50% compared with conventional techniques but was not significant in the remaining 8 types of examination [4].
We recommend that non-specialist X-ray departments should assign responsibility for paediatric imaging to a select group of interested radiologists and radiographers, thus reducing any "learning curve" effects. Optimizing examination technique is of utmost importance. Pulsed fluoroscopy is an effective method of reducing dose and this feature is now standard on modern machines. Our fluoroscopy unit defaults to 15 pulses s–1 but by selecting 3 pulses s–1 the dose is immediately reduced by a factor of five assuming the screening time is unchanged, although the image will be more "steppy" due to misregistration. (On our unit the decrease in dose is directly proportional to the decrease in pulse rate, but on some units the pulse length increases slightly as the pulse rate decreases, which would therefore result in a less marked overall fall in DAP). Our current practice is to perform most screening at 3 pulses s–1, increasing the rate if the child is very mobile or uncooperative. Images tend to be grabbed from the digital system, reserving occasional full exposures for delineation of fine detail, in difficult cases, or if there are unexpected findings. We choose to use an overcouch tube whilst maximizing the distance between the tube and the image intensifier, with the table as low and close as possible to the image intensifier. Whilst this arrangement may slightly increase the dose we have found that paediatric patients find this arrangement much less frightening and this, combined with easier access to the child for the operator and the holders, reduces the length of the examination thereby reducing dose overall. Coning to a small field of view is achieved by the operator using a light beam diaphragm for guidance and we would consider this essential for a paediatric population. We use a low attenuation carbon fibre table. A removable grid is available, and is generally only used on patients over the age of 8 years unless a younger child was particularly large for their age. Younger patients do not require a grid and this should not be used. We would not advocate the use of a unit with a fixed grid for use on children. We use 0.3 mm of added copper filtration and this is left permanently in place. Operators are encouraged to use the median values as appropriate "targets" during an examination.
DAP meters are particularly useful for assessing and comparing the dose from screening procedures as "dose x area" provides a more useful indication of overall patient exposure than measurements of entrance surface dose at different locations [13]. Fluoroscopic screening time is of limited use as a measurement of dose as it makes no allowance for the influence of dose rate or field size and to the contribution of any spot/grabbed images.
NRDs are limited in having to take into account the variation in equipment and film–screen combinations and as such give an idea of national practice, which is not necessarily the best practice. The use of local audit and critical review to ensure best practice should not be underestimated. Cook and co-workers [15] described reductions in ESD of greater than 50% for routine paediatric examinations.
Whilst third quartile data is useful in setting National Reference Doses, and hence National DRLs, their application to establishing local DRLs is of limited value. The Joint Working Party of the IPEM/BIR/RCR/NRPB/CoR [16] promotes the use of the mean of the distribution of room mean doses within an organization as the local DRL. However, an alternative method of setting a local DRL in situations were the sample size is small may be to display data as a histogram of DAPs for set age ranges and to exclude outliers beyond the limit of the curve using a mathematical "best fit" or agreed mathematical model. The DRL may then be a better reflection of "good and normal practice". It would be practical to choose a DAP based on the shape of the curve that would reliably include "normal practice" but would flag up the high dose outliers. This method would depend on having sufficient values to plot a meaningful histogram and the development of a reliable mathematical model.
For everyday working practice, established median values may give a better impression of typical local DAP values and are a useful way for an operator to compare their cases against those of their colleagues. We display median and third quartile values for all tests in our fluoroscopy room as a guide.
|
Conclusions
|
|---|
We have collected data for 2658 paediatric fluoroscopic examinations. Analysis of the data, and the derivation of third quartile DRLs, shows that our practice uses very much lower DAPs (and therefore DRLs) than the NRDs published by the NRPB and those published elsewhere in the literature. The variation between data in our study and the NRDs suggests that if the NRDs are used as a sole guide, many institutions will be falsely reassured and may be using greater doses than necessary. Only strict attention to technique and critical review of local reference levels will ensure best practice. NRDs or national DRLs, however they are determined, can only reflect "good and normal practice" if current DAP data is submitted regularly to the NRPB or other national/international bodies.
Received for publication October 5, 2004.
Revision received August 8, 2005.
Accepted for publication August 15, 2005.
|
References
|
|---|
- National Radiation Protection Board. Occupational, public and medical exposure, Documents of the NRPB, Vol 4, No. 2. Chilton, UK: National Radiation Protection Board, 1993
- United Nations Scientific Committee on the Effects of Atomic Radiation. Sources, effects and risks of ionising radiation, UNSCEAR 2000 Report, Vol. II: effects. New York, NY: United Nations, 2000
- Department of Health. The Ionising Radiation (Medical Exposure) Regulations 2000. London: Department of Health, 2000
- National Radiation Protection Board. Doses to patients from medical X-ray Examinations in the UK – 2000 Review, NRPB-W14. Chilton, UK: National Radiation Protection Board, 2002
- National Protocol for Patient Dose Measurement in Diagnostic Radiology. IPSM/NRPB/CoR. Chilton, UK: National Radiological Protection Board, 1992
- Montgomery A, Martin CJ. A study of the application of paediatric reference levels. Br J Radiol 2000;73:1083–90.[Abstract]
- Council Directive 97/43 Euratom of 30 June 1997 on health promotion of individuals against the dangers of ionising radiation in relation to medical exposure. Official Journal of the European Communities, 1997
- Hart D, Wall BF, Shrimpton PC, Dance DR. The establishment of reference doses in paediatric radiology as a function of patient size. Radiat Prot Dosim 2000. 90:235–8.
- Hart D, Wall BF, Shrimpton PC, Bungay DR, Dance DR. Reference doses and patient size in paediatric radiology. NRPB – R318 November 2000
- Martin CJ, Farquhar B, Stockdale E, Macdonald S. A study of the relationship between patient dose and size in paediatric radiology. Br J Radiol 1994;67:864–71.[Abstract]
- Council of the European Communities. European guidelines on quality criteria for diagnostic radiographic images in paediatrics, EUR 16261. Luxembourg: Office for Official Publications of the European Communities, 1996
- Chapple CL, Faulkner K, Lee REJ, Hunter EW. Results of a survey of doses to paediatric patients undergoing radiological examinations. Br J Radiol 1992;65:225–31.[Abstract]
- Chapple CL, Faulkner K, Lee REJ, Hunter EW. Radiation doses to paediatric patients undergoing less common radiological procedures involving fluoroscopy. Br J Radiol 1993;66:823–7.[Abstract]
- Servomaa A, Komppa T, Heikkila M, Parviainen T. Patient doses in paediatric fluoroscopic examinations in Finland. Radiat Prot Dosim 2000;90:239–43.[Abstract]
- Cook JV, Kryriou JC, Pettet A, Fitzgerald MC, Shah K, Pablot SM. Key factors in the optimisation of paediatric x-ray practice. Br J Radiol 2001;74:1032–40.[Abstract/Free Full Text]
- Guidance on the Establishment and Use of Diagnostic Reference Levels for Medical X-Ray Examinations. IPEM Report 88. York, UK: Institute of Physics and Engineering in Medicine, 2004
This article has been cited by other articles:
|
|
|
|
 
W Mazrani, K McHugh, and P J Marsden
The radiation burden of radiological investigations
Arch. Dis. Child.,
December 1, 2007;
92(12):
1127 - 1131.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
BJR review of the year - 2006
Br. J. Radiol.,
March 1, 2007;
80(951):
147 - 151.
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
