| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Correspondence |
I would like to comment on two papers published recently in the British Journal of Radiology; a paper published by Paisley et al in the December 2004 issue [1] and one by George et al in the October 2004 issue [2]. Both papers incidentally, in my opinion, represent excellent work.
The paper by Paisley et al in the December 2004 issue presents results of radiation dose measurements during invasive cardiac procedures categorised by clinical code [1]. This paper, I believe, highlights the present state of confusion that exists in respect of patient dose assessment in diagnostic radiology.
The paper presents measurements undertaken by means of a dose–area product (DAP) meter and refers to the results as "patient doses". Unfortunately, this is not correct, since DAP meters do not measure absorbed dose to a specified medium that is a defined term according to international standards. The DAP meter combines two quantities that have a fundamental bearing on the quality of an X-ray examination, namely dose in air (related to entrance surface dose (ESD)) and field size, into a single quantity. Within the combined quantity, DAP, neither of the individual quantities is known with any degree of accuracy, unless specific further information concerning the X-ray examination is also recorded. This includes field size at the patient entrance surface and/or X-ray focus to skin distance. Thus the DAP meter reading on its own actually lowers the quality of physical data captured during an X-ray examination through the resulting uncertainty. Such an action would not be permitted in any other branch of physical measurement. If ESD and field size are desirable quantities in their own right, then in order to describe adequately the X-ray imaging technique they should be assessed individually.
The effects of the uncertainty caused by this combination of factors may be highlighted by considering the results of patient dose measurements (both DAP and ESD) obtained as part of the National Survey of Doses to Patients undertaken in England by the NRPB in 1985 [3]. In the resulting report [NRPB – R200] results are presented for both male and female patients separately. For the PA chest examination, the mean ESDs for male and female patients were 0.26 mGy and 0.2 mGy, respectively (ratio 1.3). For the DAP values the male and female mean figures were 0.6 Gy cm2 and 0.32 Gy cm2, respectively (ratio 1.87). Thus, is the increased ratio for the DAP values compared with the ESD values due to an, on average, larger field size employed for male patients compared with females, or higher ESD for the same field size? In either case there appears to be scope to optimize practices for male and female patients separately, but the results taken individually do not indicate how. The above finding for male and female patients is not confined to posteroanterior chest examinations.
In a paper by George et al in the October 2004 issue, results are presented on patient dose optimization in plain radiography based on standard exposure factors [2]. The method employed by the authors does not appear in the UK National Protocol for Patient Dose Measurements in Diagnostic Radiology, even though the use of calibrated X-ray outputs as a basis for computing patient dose has formed the basis of radiotherapy applications for many years. Modern X-ray generators have demonstrated a high degree of stability for some time now, making the method eminently practical. Thus the paper raises questions about what is an acceptable dosimetric method as well as what is meant by optimization.
Patient dose measurements and the establishment of appropriate Diagnostic Reference Levels (DRLs) are recommended in the 1997 EC Directive 97/43 EURATOM under Article 4 dealing with optimization [4]. Unfortunately, the interpretation of optimization is often confined only to the assessment and comparison of risk from ionizing radiation. This philosophy has led to the implementation of the concepts of effective dose and energy imparted to patients who undergo an X-ray examination. These approaches map an ESD value for an individual patient on to average organ risk weighting factors in order to derive an equivalent risk related dose. Such a quantity is then used to compare the risks from different types of examinations.
Unfortunately, such an approach ignores the benefits arising from an X-ray examination. This is represented by the image quality and corresponding diagnostic outcome derived from the imaging procedure. Since complete optimization is concerned with maximization of risk/benefits, the outcome is relevant. Thus a totally risk driven optimization process must assume that all X-ray examinations have a constant diagnostic outcome.
The clinical outcome from individual diagnostic procedures, however, cannot be normalized, in that the image of a chest is different from that of an abdomen etc. Also, each patient's diagnosis is unique to that patient. Because anatomical, structural noise plays a major role in determining the detectability of abnormalities (pathology), it is highly unlikely that diagnostic outcomes are constant for all types of examinations, given the wide range of structures involved. Hence when viewed from a risk only point of view, the risk/benefit optimization equation is unbalanced. Indeed, much effort is now being expended to reduce the effects of structural, anatomical noise, through the use of subtraction or three-dimensional imaging methods based upon digital radiographic technology.
Can all of these different methods, which will obviously provide different diagnostic outcomes, continue to be compared solely on the basis of risk? The goal posts have already been moved towards an increase in clinical information arising from an examination, even if this involves an increased dose burden. Such an approach aims at producing a reduction of clinical risk through improvement in the diagnostic accuracy provided by a procedure. Justification should be based upon accuracy of diagnosis as well as dose considerations. The proposed deployment of volume CT as a screening tool is a fine example of this dilemma. The information genie has already been released from the risk/dose bottle.
Confusion in the field of patient dose assessments is a significant problem internationally. First, because there is no universally accepted understanding of what is meant by optimization. Second, since the International Commission on Radiological Protection recommended the use of diagnostic reference levels for patients, the scientific community has spawned a wealth of terms and nomenclature. Such terms encountered in publications, guidance documents etc. include [5]:
As well as the wealth of terms employed, a wide variety of dose quantities have been employed:
ESD, entrance surface air kerma (ESAK), entrance skin exposure (ESE), DAP, entrance dose per radiograph, DAP per examination.
Periapical entrance dose (PED), dose–width product (DWP)
ESD rate, ESAK rate, peak skin dose, cumulative dose, DAP
CT dose index (CTDI), weighted CTDI (CTDIw), Z – averaged weighted CTDI (CTDIvol), multiple scan average dose (MSAD), dose–length product (DLP)
ESD, ESAK, average glandular dose (AGD), mean glandular dose (MGD).
The situation outlined for diagnostic radiology would not be tolerated in the field of radiotherapy. Treatment-dependent dose quantities which could not be intercompared or readily understood by medical staff who prescribe treatments have long been eliminated. Indeed, the accurate quantification of dose is the very foundation of the application of scientific methods in the field of radiotherapy, since the assessment of clinical outcomes (cure rates etc.) fundamentally depends upon dose delivered and its distribution. Why, then, is it tolerated in diagnostic radiology? The most likely explanation lies in the lack of a clear understanding of the meaning and relevance of the term "optimization" in the field of diagnostic radiology. For example, the employment of a single DRL for both male and female patients has not, to the best of my knowledge, ever been justified to the radiological community as the only, most realistic or appropriate approach to the optimization of practices.
In the field of diagnostic radiology the measurement of dose underpins all scientific investigations, including the development of a framework for assessing quantitatively the diagnostic outcome from X-ray examinations. Such developments are already underway in the field of Computer Aided Diagnosis (CAD). The international scientific community must, therefore, come together to eliminate the present state of confusion that exists in the field of patient dose measurement in diagnostic radiology. Such methods must provide rigorous scientific measurements, which can be readily understood and intercompared, together with diagnostic outcomes, across the whole range of radiological examinations including CT, mammography, radiography and fluoroscopy. Without a proper framework for optimization, such concepts as justification, net benefit, total potential diagnostic benefit and efficacy cannot be assessed and/or compared. These are all terms employed in the Ionising Radiations (Medical Exposure) Regulations [IR(ME)R] 2000 [6], concerned with legal requirements for protection of the patient from the medical use of ionizing radiation.
In a recent study published in the British Medical Journal [7] it was shown that there is a general ignorance within the medical profession concerning the levels of radiation employed in diagnostic radiology. Given the fact that over 300 000 manSv of radiation are prescribed annually in diagnostic radiology in Europe, the scientific community has a professional responsibility to create an environment whereby this ignorance can be eliminated. An important part of this process must involve collaboration amongst appropriate groups within the radiological community, in order to develop a framework for the quantification of diagnostic outcome and therefore clinical benefit. In this way, a wide range of relevant scientific initiatives within the field of diagnostic radiology can be placed upon a firm scientific footing.
Yours etc.
Co-ordinator, RADIUS Group, EC Radiation Protection Research Programme
Received for publication January 12, 2005. Revision received April 6, 2005. Accepted for publication April 20, 2005.
References
This article has been cited by other articles:
 |
|
 |
|
  B F Wall Response to "Radiation dose measurement and optimization". Br. J. Radiol., April 1, 2006; 79(940): 356 - 357. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| BJR | DMFR | IMAGING | ALL BIR JOURNALS |
