Criticism of "absorbed dose"
Two letters from the European Radiation Protection Research newsletter
January 1999

"Absorbed dose is not valid for the specification of biological effects.
Reply to Pascal Pihet and Hans Georg Menzel (see Issue nr 3, July 1998)

Arguments against the validity of absorbed dose as a quantifier for the biological effects of ionizing radiations have been broached since the early 1970's. In these past 25 years or so we have seen the ICRP system of radiological protection evolve into a muddle of complexity challenged by ever-increasing criticism of the use of absorbed dose (e.g. Simmons, Cameron, Japanese, Sabol).
During these years, ICRU has given impressive guidance on quantities and units all of which have been thought in one way or another to be relevant to the determination and measurement of radiation effects. Yet in practical application, operational protectionists confined by the regulatory aspects, are near despair; radiotherapists and radiobiologists, working mainly with computed dose distributions and empirical RBEs, simply ignore the formal quality parameters whilst radiation bio-physicists have been rejoicing at the widening horizons for investigation in an ever-expanding universe of dose-based uncertainty.
Simulation models intended to describe the mechanisms of radiation damage occur literally by the dozens yet still we have not agreed on how best to predict the actual biological effectiveness of a known radiation field with tolerable accuracy without resorting to empirical techniques even for the simplest of mammalian cells. Amid this chaos, the validity of the fundamental concepts must surely be suspect, even if only intuitively?
To make a realistic prediction of damage using absorbed dose, we have to resort to clever tricks involving the relative dose-modifying quantities such as measured RBEs or artificial radiation weighting factors, depending on application. Among the quality parameters, LET still enjoys recognition despite well-founded criticism of its use. LETs, restricted in energy cut-off and the analogous microdose quantity, lineal energy, fare somewhat better but their use still does not adequately account for radiation quality to enable prediction of bio-effectiveness. Almost without exception. all the attempts to predict the effectiveness of radiation fields involve the use of energy transfer quantities and are found to be seriously inadequate.
It is timely that Pihet and Menzel challengingly make their moot point commenting in justification of dose. Certainly dose may be accepted as a fundamental quantity in dosimetry but it is not truly the fundamental quantity in energy transfer processes. The primary collision rate along the charged particle track, i.e. the total collisional interaction cross-section, has that honour. This becomes more explicit when the interaction cross-section is written as the zeroth moment of energy transfer whereas the LET is the weighted first moment. If LET, or its related restricted quantities including the stochastic lineal energy, is used to express quality then compatibility of units dictates that charged particle fluence should be used as the field quantity. Under charged particle equilibrium conditions this gives effectively the dose, a measurement of mean energy imparted per unit mass of target material. On the other hand if the cross-section for induction of the bio-physical effect is used for quality, the relevant field quantity is still the particle fluence. However, in the latter case, the product represents the absolute mean number of initial damaging events in the target of interest - the amount of energy transferred is not necessarily relevant.
Conceptually there exists a crucial difference between the two methods of damage specification. Indeed, it can be demonstrated empirically and theoretically that the bio-effect cross-section is best expressed independently of energy transfer parameters.
Our colleagues Pihet and Menzel remind us that absorbed dose has the properties of: scientific rigour; is easily and accurately measurable: standards are available, etc. This is so, but in no way do these attributes exclude change in choice of parameters or units for the main ICRU quantities. Introduction of additional new quantities is unlikely to pose insurmountable difficulties in this respect.
With regard to the claimed successful application of treatment regimens in radiotherapy, one should remember that these are based on absorbed dose distributions, usually calculated but monitored by measurement at a pre-selected position. No measurement of the actual physical or biological "quality" distribution at the point of interest is carried out. Indeed, the clinician does not yet have at his disposal any instrumentation to measure the actual bio-effectiveness in internal or external therapy.
Intercomparisons of "dose" tell nothing about quality. Use of dose alone for measurement of the effectiveness of electrons implicitly assumes that the bio-effect per unit dose is independent of the LET which is manifestly not true. It so happens that change in the bio-effect produced by electrons for equal doses (differing energy spectra) can be as much as 50%. Differences among the typical quality parameters in use are less obvious for electrons than for heavy particles. For electrons, no matter how accurately dose may be measured at best it can serve only as a very crude estimate of the bio-effectiveness. Despite the existence of detailed protocols, a rigorous description of the basic damage mechanism is missing along with the desired instrumentation. Currently, microdosimeters probably provide the best information on the physical quality of the radiation field of interest which, although the events are stochastic, is not too dissimilar from those for non-stochastic LET spectra when weighted for the mean chord distributions in a sphere. Nevertheless, it has not yet been demonstrated that the energy-based micro-quantities can be related to the desired quantity, the `biological effectiveness' of the radiation. Again, all of these inadequacies brings the validity of the conventional dosimetry quantities into question.
The proposed `proof' that absorbed energy is invalid as a predictor of radiation effects rests on three main factors:
1) the demonstration that bio-effect cross-section data for all radiation types can be best correlated into a unified curve when plotted as a function of the mean spacing between primary interactions for the relevant charged particle tracks.
2) the physical proof that there are multiple radiosensitive sites each of ~2nm dimensions within the cell nucleus and
3) the demonstration that individual electron tracks have very low efficiency at causing inactivation of mammalian cells (~1 per 105 delta rays from heavy particle tracks). As delta rays have negligible effect it follows from well-established stopping power theory that the initial damage action must be controlled by the interaction cross-section and not by the LET or related quantities and hence it is concluded that the bio-effectiveness does not depend on energy transfer.
Consequently LET, and therefore absorbed dose, does not play a role in the inactivation mechanism.
If absorbed dose is agreed to be not only inappropriate but actually invalid for describing radiation effects then priority action, possibly in EC Fifth Framework programme, should by taken to examine the consequences for the future.

David E.Watt
School of Physics and Astronomy
University of St. Andrews
St. Andrews KYI6 9SS UK.
8 September 1998

Is absorbed dose always appropriate to quantify biological effects induced by ionising radiation?
Reply to Pascal Pihet and Hans Georg Menzel (see Issue nr 3, July 1998)

The recent article by Pihet and Menzel on the importance of the concept of absorbed dose in radiation protection was thought provoking. We agree with the author in interpreting that the underlying assumptions apply particularly to radiotherapy calculations involving a defined mass of homogenous cells and receiving a relatively large dose. Although it is implicit in weighting factors that different organs respond differently to the same absorbed dose, there is no convenient method of integrating weighting factors to predict a whole body response. We also support the authors' suggestion that the same criteria of absorbed dose do not apply to low level radiation protection issues. Low level irradiation of a whole animal may have absolutely no biological effect, may consequentially improve the health of the animal ( hormesis or diagnosis), or may disimprove the health of the animal. At low doses no biological effects will be directly proportional to dose because of the multiplicity of confounding variables (cell - cell interactions, co-operative repair, induced instability, bystander and non-targeted effects etc.).
The lack of contemporaneousness between low level radiation exposure and the expression of any damage allows the possibility of other modifying substances to exert confounding or synergistic influence. In the absence of a unique marker of radiation, it becomes impossible to ascribe any effects uniquely to the radiation. It is our opinion that there is a biological threshold dose below which knowledge of the absorbed dose is irrelevant to predicting the biological response. The dose at which this threshold occurs is unlikely to be the same for all species or even individuals within that species.

Colin Seymour and Carmel Mothersill
Dublin Institute of Technology
Kelvin Street,
Dublin 8, Ireland
13 September 1998

Extract from Low Level Radiation Campaign response to DETR consultation on the recycling of contaminated materials arising from Clearance of nuclear sites:
"...radiation causes genetic damage by discrete tracks each of which impinges only on the cells within the range of the charged particle and only on the cells which lie in line with the direction of the particle (The recent discovery of an as yet poorly understood "bystander effect" does not invalidate this statement - it merely supports the contention that knowledge of the mechanisms of radiation-induced genetic mutation is very inadequate). On the microscopic level (i.e. at the level of the cell) radiation "dose" or "energy transferred" is very high for those cells which are hit, and zero for those which are missed.
The conventional approach of averaging the energy transfer from radioactive decay events across a whole organ or the entire body is like emptying a Colt 45 into a football stadium and averaging the effects of the 6 bullets across all the 25,000 spectators. The assumption that between them 25,000 people should be able to stop six bullets without any of them feeling more than a tap on the arm will not console the six grieving families."

If you are seeing this page full screen (i.e. without a navigation bar on the left) you can't see how the rest of the site is organised.
This Home page link takes you to the index page, which has links to all the topics we discuss on the site [only use it if this page is full screen]

Send email to: with questions or comments about this web site.

This page was last updated May 2001