Chapter 1
APPENDIX I

The conditions of exposure during CT examinations are quite different from those in conventional x-ray procedures and specific techniques are necessary in order to allow detailed assessment of patient dose from CT. National surveys of CT practice using such methods of dosimetry have established the increasing importance of CT as a significant source of medical x-rays for populations in developed countries (1). Evidence from dose surveys has also indicated potential scope for improvement in the optimisation of protection for patients undergoing CT and the need for more widespread assessment of typical levels of patient dose as part of routine quality assurance (2,3). Inherent differences in the design of CT equipment lead to variations between scanner models by up to a factor of three in the calculated values of effective dose for standard examinations under conditions of similar image quality (4). However, larger variations in dose are apparent in clinical practice, with the minimum and maximum values of typical dose for a given type of procedure varying by factors, for example, of 10-40 in the UK (4) and 8-20 in Norway (5); this is largely as a result of differences in the local scanning technique typically employed for a particular type of examination, as determined by the number and thickness of slices imaged, the couch increment between slices, the use of contrast medium for additional scans and the exposure settings selected.

The Examples of Good Imaging Technique given in the Lists of Quality Criteria are intended to help avoid unnecessary exposures in CT. The Criteria for Radiation Dose to the Patient indicate diagnostic reference dose values for general types of examination as a practical means of promoting strategies for optimisation of patient protection. The purpose of a reference dose quantity for a diagnostic medical exposure is to provide quantification of performance and allow comparison of examination techniques at different hospitals. Diagnostic reference dose values should not be applied locally on an individual patient basis, but rather to the mean doses observed for representative groups of patients. Reference dose values are intended to act as thresholds to trigger internal investigations by departments where typical practice is likely to be well away from the optimum and where improvements in dose-reduction are probably most urgently required. Typical levels of dose in excess of a reference dose value should either be thoroughly justified or reduced. In general, patient doses should always be reduced to the lowest levels that are reasonably practicable and consistent with the clinical purpose of the examination.

The derivation of the diagnostic reference dose values is described in Chapter 2. Reference dose quantities and methods for their assessment are discussed below.

The principal dosimetric quantity used in CT is the computed tomography dose index (CTDI). This is defined (6) as the integral along a line parallel to the axis of rotation (z) of the dose profile (D(z)) for a single slice, divided by the nominal slice thickness T:

(mGy)   (1)

In practice, a convenient assessment of CTDI can be made using a pencil ionisation chamber with an active length of 100 mm so as to provide a measurement of CTDI₁₀₀ expressed in terms of absorbed dose to air (mGy). Such measurements may be carried out free-in-air on or parallel with the axis of rotation of the scanner (CTDI_{100, air}), or at the centre (CTDI_{100, c}) and 10 mm below the surface (CTDI_{100, p}) of standard CT dosimetry phantoms. The subscript `n' (_nCTDI) is used to denote when these measurements have been normalised to unit radiographic exposure (mAs). Further discussion of the quantity CTDI is given in Chapter 2.

Such measurements of CTDI in the standard head or body CT dosimetry phantom may be used to provide an indication of the average dose over a single slice for each setting of nominal slice thickness. On the assumption that dose in a particular phantom decreases linearly with radial position from the surface to the centre, then the normalised average dose to the slice (7) is approximated by the (normalised) weighted CTDI (CTDI_w):

(mGy(mAs)-1)   (2)

where C is the radiographic exposure (mAs) and CTDI_100,p represents an average of measurements at four different locations around the periphery of the phantom. Values of _nCTDI_w can vary with nominal slice thickness, particularly for the narrowest settings.

Two reference dose quantities are proposed for CT in order to promote the use of good technique:

(mGy)   (3)

where _nCTDI_w is the normalised weighted CTDI in the head or body phantom for the settings of nominal slice thickness and applied potential used for an examination (Equation 2) and C is the radiographic exposure (mAs) for a single slice in serial scanning or per rotation in helical scanning.

Monitoring of CTDI_w for the head or body CT dosimetry phantom, as appropriate to the type of examination, provides control on the selection of exposure settings, such as mAs.

(mGy cm)   (4)

where i represents each serial scan sequence forming part of an examination and N is the number of slices, each of thickness T (cm) and radiographic exposure C (mAs), in a particular sequence. Any variations in applied potential setting during the examination will require corresponding changes in the value of _nCTDI_w used.

In the case of helical (spiral) scanning:

(mGy cm)   (5)

where, for each of i helical sequences forming part of an examination, T is the nominal irradiated slice thickness (cm), A is the tube current (mA) and t is the total acquisition time (s) for the sequence. _nCTDI_w is determined for a single slice as in serial scanning.

Monitoring of DLP provides control on the volume of irradiation and overall exposure for an examination.

Procedures for estimating CTDI_w and DLP are given below.

Comparison of performance against the criteria for each particular type of examination requires assessment of the values of the reference dose quantities associated with the parameters of technique typically used when scanning a standard-sized adult patient. In the absence of a well- defined scanning protocol, typical dosimetric practice should be determined on the basis of the mean results derived for a sample of at least 10 patients for each procedure.

CTDI_w may be assessed directly from Equations (2) and (3) using the results of measurements of CTDI_{100, p or c} for the head or body CT dosimetry phantom carried out during routine performance testing. Such measurements may be accomplished using thermoluminescent dosemeters (TLDs) or more conveniently using an appropriately calibrated 100 mm long pencil-shaped ionisation chamber (8). It has been recommended by the International Electrotechnical Commission that values of CTDI_w should be displayed on the operator's console of the CT scanner, reflecting the conditions of operation selected, although an appropriate correction should be included if the nominal slice thickness is not equal to the couch increment per tube rotation (9). Typical values of _nCTDI_w for a wide range of scanner models have been collated into a reference database on CT dosimetry that has been published on the Internet (10). Some standard dose data for a selection of scanners is given, for illustrative purposes, in Appendix I to Chapter 2.

Estimates of CTDI_w may also be made using the typical dose data commonly provided by manufacturers in fulfilment of the requirements of the Food and Drug Administration (FDA) in the USA. Accordingly, manufacturers of CT scanners are obliged to report values of CTDI measurements in the standard head and body CT dosimetry phantoms using a specific protocol (11) for which there are important differences from the approach advocated in this report; such values of CTDI_FDA refer to an integration length equivalent to 14 nominal slice thicknesses (rather than 100 mm) and are expressed in terms of absorbed dose to PMMA (rather than air). Similar measurements have previously been recommended by the International Electrotechnical Commission (IEC) as part of constancy testing in CT (12). However, values of CTDI_FDA determined in the phantoms will be only slightly less than CTDI₁₀₀ for the largest settings of slice thickness, but more significantly so for smaller slice thicknesses. Table 1 gives broad factors (13) to allow the estimation of CTDI_w from such manufacturers data (CTDI_FDA).

As a practical alternative, estimates of CTDI_w for the head or body CT dosimetry phantom may be derived from simpler measurements of CTDI made free-in-air (CTDI_air) under similar conditions of exposure (H = head, B = body):

(mGy)   (6)

where the factor P_{H or B} is given by:

(7)

(8)

Measurements of CTDI_air are easily accomplished with either the 100 mm pencil-shaped ionisation chamber or a shorter length of TLDs since the tails on the dose profiles in air are less significant than in a phantom in view of the lower amount of scattered radiation. Some typical values of the factor P for selected scanner models are given in Appendix I to Chapter 2. Further data for a wider range of models are available in the reference database on CT dosimetry (10).

Subsequent estimates of DLP for an examination may be derived using Equations (4) and (5), with knowledge of appropriate values of _nCTDI_w for the scanner and details of the particular scanning protocol used. In the case of examinations involving separate scanning sequences in which different technique parameters are applied (such as slice thickness or radiographic exposure, for example), the total DLP should be determined for the entire procedure as the sum of the contributions from each serial or helical sequence.

ASSESSMENT OF EFFECTIVE DOSE
In addition to comparison of performance against reference dose values, there is sometimes a need to assess effective dose (14) for CT procedures so as, for example, to allow comparison with other types of radiological examination. The effective dose for a particular scanning protocol may be estimated from a measurement of CTDI_air utilising scanner-specific normalised organ dose data determined for a mathematical anthropomorphic phantom using Monte Carlo techniques (15,16). For types of scanner not included amongst these calculations, appropriate data sets may be selected from those available on the basis of similarity of values of P (Equations (7) and (8)) (17,18).

Alternatively, broad estimates of effective dose (E) may be derived from values of DLP for an examination using appropriately normalised coefficients:

(mSv)   (9)

where DLP (mGy cm) is the dose-length product as defined in Equations (4) or (5) and E_DLP is the region-specific normalised effective dose (mSv mGy^-1 cm^-1).

General values of E_DLP appropriate to different anatomical regions of the patient (head, neck, chest, abdomen or pelvis) are given in Table 2.

Such an estimate of effective dose may also be derived from a measurement of CTDI_air on the basis of Equation (6) and Equations (4) or (5) to determine DLP.