Continuing developments in the technology and clinical application of CT are ensuring not only its increasing role in population healthcare, but also its attendant position as a special focus for efforts in radiation protection.
This document seeks to provide welcome updated guidance to help establish best practice for patient protection in MDCT. It contains useful descriptions of the new technology inherent in MDCT, although this seemingly early (yet public) draft would benefit from further editing to eliminate typographical errors and unnecessary repetition of material, improve clarity of expression and the uniform application of a concise style, and provide a better focus for the presentation of practical recommendations on technique and dosimetry for those involved in the clinical use of contemporary MDCT. Unfortunately, the present text often fails to reflect adequately the section headings and includes, perhaps, undue emphasis on SSCT and older 2- and 4-MDCT systems (rather than contemporary 16- or 64-MDCT systems). There is also inadequate coverage of significant developments in Europe and existing international advice, recommendations and regulations for managing patient dose in CT (in relation, for example, to reference doses). The document should conclude with a short section summarising the advice and recommendations being presented.
1. Table of Contents, lines 27-92
There is a mismatch between many of these section headings and their present content.
Section 3.3 has the same heading as Section 3 (Operator choices that affect patient dose).
Section 4 might be better called ‘Dose management in clinical practice’
Section 4.3 does not include information on risk and might be called ‘Technique and dose for particular CT examinations’.
2. Summary points, lines 96-173
This is presently quite long and detailed. It might usefully focus in a more concise manner on the practical information necessary for effective management of patient dose in MDCT.
Review, for example, the merit of including the particular points at lines 129 and 134.
Avoid jargon (such as “pretty pictures” and “one size fits all”).
3. Section 1.2, line 227
Do MDCT scanners use 3rd (line 227 or 4th (line 282) generation CT geometry?
4. Section 1.2, line 230-231
The original EMI scanner simultaneously acquired 2 images.
5. Section 1.3, Fig 1.1, lines 267-275
There is some confusion and inconsistency in the present description of detector arrays for SDCT and MDCT. Surely Figure 1.1 shows detector arrays having 1 row and 16 rows, each with 900 elements?
6. Section 1.4, line 381-386
Does the report provide clear answers to these questions?
7. Section 2.3.1, lines 522-523
Review the intended meaning of this unclear sentence.
8. Section 2.3.2, line 544
Clarify the term ‘long volume’.
9. Section 2.4, Dose surveys and reference levels
This section fails to define the purpose and use of reference doses and is thus confusing.
There is no reference to the 2004 European Quality Criteria for Multislice CT (http://www.msct.eu/CT_Quality_Criteria.htm#Download%20the%202004%20CT%20Quality%20Criteria). These update EUR 16262 (1999).
The reference for Shrimpton et al (line 549) should be 2005, not 2006.
It is unclear from Table 2.4 exactly what doses are being compared (mean values?) and to what effect. Those from Shrimpton et al 2005 represent 75th percentile values of the observed national dose distributions (including a mix of scanner technology). Separate results are presented by these authors for SSCT and MDCT, together with corresponding UK national reference doses.
The section does not include any data for paediatric CT (which was previously flagged in Section 1.4 as being of particular importance within the Document).
Clarify what is meant by ‘are on the higher side’ (lines 566-569).
10. Section 2.5, Perspectives on radiation risks, lines 591-625
It would be helpful to include some approximate numerical data on risks in these 2 sub-sections, in addition to the qualitative statements.
For example, include at line 593 the range of typical maximum skin doses from CT examinations and the threshold for skin injury.
The final sentence of the paragraph at lines 601-2 begs the question (and further advice) of what operators should do to help avoid such injuries.
Similarly, some quantitative indication is required of the ranges of stochastic risks for CT on adults and children.
11. Section 2.6, Responsibilities for patient dose management
This section fails to mention the operator (patient’s physician) as being ultimately responsible for examinations on individual patients.
The section at lines 649-654 should clarify the role of reference doses (as a practical aid to the optimisation process) in helping limit unnecessary patient dose.
Is the statement at lines 658-659 really true (and the reference to Section 3.1 correct)? Early CT dose surveys in the 1980’s had already highlighted for the radiology community the relatively high levels of patient dose and significant variations in practice.
12. Section 3.1
This section should include reference to the criteria on image quality and dose developed for CT by the European Commission (1999 and 2004: http://www.msct.eu/CT_Quality_Criteria.htm#Download%20the%202004%20CT%20Quality%20Criteria).
13. Section 3.1.2, line 757
What is meant by ‘a standard dose CT’? Surely, all doses should be as low as reasonably practical for the intended purpose of the examination?
14. Section 3.1.3
The purpose of this section is unclear from the heading and the content.
Mention is made at line 762 of the considerable efforts of the EC in relation to optimisation of patient protection, although no references are included (such as reports EUR 16260, 16261, 16262, and MSCT 2004: http://www.msct.eu/CT_Quality_Criteria.htm#Download%20the%202004%20CT%20Quality%20Criteria).
15. Section 3.2
Whereas some discussion of SSCT and 2- and 4-MDCT systems provides useful background information, the main focus of the document should perhaps be more on contemporary CT (such as 16-MDCT and beyond since these systems will, as mentioned previously in the Document, soon begin to dominate practice).
The sentence at line 846 requires at its end the additional clarifying text ‘for the same anatomical coverage’.
16. Section 3.3, Operator choices that affect patient dose
This sub-section has the same heading as Section 3. Its intended content is unclear since the material in, for example, Sections 3.4, 3.5 and 3.6 would also fit within the present heading.
17. Section 3.3.2, line 961
The term ‘dose reduction’ should be replaced by ‘mAs reduction’, since, for example, the effective doses to an adult and a child will be different even for the same mAs (and so a change in mAs does not lead to a proportionate change in effective dose for different sized patients).
18. Section 220.127.116.11, line 963
Is 80kg now a standard adult (in the USA)?
19. Section 18.104.22.168, Table 3.1
Would it be clearer to replace the term ‘neurological CT’ in the heading (and elsewhere in the text) by ‘head CT’?
Do the factors shown also apply to circumstances where the image increment is not equal to the image width?
20. Section 22.214.171.124
Some clarifications are required concerning the ‘dose reductions’ quoted at lines 991, 992, 1033 and 1068. Do these data refer, for example, to organ doses, effective doses, localised doses or whole body doses?
Is the reference at line 1041 to ICRP 1991 correct?
21. Section 3.3.4, line 1188
Clarify ‘reduce dose by up to 50%’ (see point 20, above).
22. Section 3.4
At line 1212, clarify ‘37% dose reduction’ (see point 20, above).
The sentence at line 1217 should be revised for clarity to ‘..if the tube current is not appropriately increased to compensate for the lower photon fluence at lower tube voltages.’ or similar.
23. Section 3.6, Scan mode
It is not clear from the heading what the focus of this section is. Scan mode often refers to the selection of sequential or helical (spiral) scanning. The material included would seem better placed in Section 3.7 on ‘Scan coverage’. Should the Document include advice on the appropriate choice (and implications for patient dose) between sequential and helical scanning?
In any case, is the statement at lines 1238-1240 correct? Or should not the phrase ‘should be avoided’ be replaced by ‘should be preferred’?
24. Section 3.7, Scan coverage and indication
The material in this section is somewhat vague.
In the final sentence, who is ‘to inform the patient’s physician’? Surely, requests for ‘exams of inappropriate anatomy or for non-medically-necessary indications’ would not be performed as part of the process of justification?
It would be helpful if the Document included further discussion and gave clear recommendations on the use of MDCT for population screening for disease and other CT procedures on asymptomatic patients.
25. Section 3.8
Could this section be re-focussed to provide clearer information and recommendations on best practice (appropriate choices) for the MDCT operator?
Line 1258: Clarify ‘Dose savings of 30%’ (see point 20 above).
At line 1299, clarify the term ‘reprojected metal traces’.
26. Section 3.9
The first sentence requires some clarification, along the lines of ‘CT images should always be obtained with the lowest radiation dose to the patient that is achievable for the clinical purpose of the examination.’.
The final sentence seems to ignore existing international recommendations (eg from ICRP) and regulations (eg Directives from the EU and Basics Safety Standards from IAEA) formalising the adoption and practical use of reference doses (diagnostic reference levels) in support of the optimisation of patient protection.
27. Section 4
This might perhaps be renamed ‘Dose management in clinical practice’.
Lines 1341-1344: Is the classification of studies as requiring low, standard or high radiation doses part of justification or optimisation?
28. Section 4.1
Line 1374: A reference is also required for the EC publication (RP118).
Lines 1385-7: Surely CT examinations are justified by the physician and not the patient? How will the proposal lead to a reduction in unnecessary CT examinations?
Lines 1391-92: If relevant in an international Document, include a reference for the US Bills of Right and Responsibilities.
29. Section 4.3
This section does not include information on risk and might be called ‘Technique and dose for particular CT examinations’.
In general, the purpose of the information presented in the Tables is not clear. Many of the columns in each table have identical headings (such as ‘Low kVp chest CT’ or ‘Low dose CT colonography’), but with differing levels of dose. What the operators of MDCT might need is a more focussed discussion on best practice for each type of procedure.
The Table headings are too long and often confusing.
Effective doses are presented with inappropriate levels of precision.
Is it appropriate (recommended by ICRP) to include separate estimates of effective dose for males and females?
Line 1439: Clarify ‘higher end scanners’.
Line 1478: Missing reference.
30. Section 4.3.2, Coronary CT
Some discussion would be useful of the role of dual source CT for coronary imaging.
Line 1518: Clarify ‘slim patients’.
31. Section 4.3.3
Line 1568: Clarify ‘two CT passes’.
32. Section 4.3.4, CT for trauma
Lines 1584-5: Review the need for this statement in the Document.
Lines 1592-3: This statement requires further clarification. What is a ‘typical trauma patient’? Does 16 mSv refer to single or multiple MDCT examinations?
The key point is surely that timely CT examinations can save lives in cases of severe injury and considerations of dose reduction might be secondary to speed of imaging and diagnosis.
Lines 1602-13: Much of this material has been covered previously elsewhere in the Document.
Line 1611: Review the use of ‘recently’ for a report published in 2003.
33. Section 4.3.6, CT guided interventions
Line 1651: Clarify “passes”.
Table 4.5A: Clarify what is meant by ‘Average effective dose (50% range)’ and the meaning of the figures in parentheses shown for the physician doses. Review precision of dose data presented.
34. Section 4.3.8, CT of the pregnant patient
Line 1695: Is this statement true for wide beam MDCT scanners?
Line 1701: Surely for all CT examinations, the scan volume should be restricted to the necessary anatomy, not just in the special case of pregnant patients?
Line 1703: Clarify “step-and-scan protocol”.
Table 4.6: Fetal doses should be presented in terms of mGy rather than cGy.
35. Section 4.4
Much work has already been completed on the suggested development of guidelines for different indications in CT, including for example those in the 2004 European Quality Criteria for MSCT (http://www.msct.eu/CT_Quality_Criteria.htm#Download%20the%202004%20CT%20Quality%20Criteria) and those by the Royal College of Radiologists in relation to Recommendations for Cross-Sectional Imaging in Cancer Management (RCR(06)1) (http://www.rcr.ac.uk/index.asp?PageID=310).
36. Appendix A, How to describe dose in CT
In general, this section fails to address how users of MDCT should assess doses for adult and paediatric patients from contemporary MDCT with wider beams than those previously used in SSCT and also AEC.
37. Section A1
Lines 1743-1753: Is the historical discussion of early (and long since outdated) FDA initiatives really relevant here, or does it merely perpetuate confusion?
Line 1782: SI units are mandatory for this international Document (with no inclusion of rad or R).
Lines 1805-8: Explain more fully the meaning and consequences of the phrase ‘occasionally the length of irradiation goes beyond the 100mm that the pencil chamber is designed’.
Provide clearer guidance on whether CTDI and CTDIvol are still valid quantities for MDCT and, if so, how they should be measured. Give further details of the ‘new chambers’ for MDCT, if these are being recommended for use. Otherwise, provide clear guidance on what dose quantities should be used for assessing practice in MDCT.
Clarify which dosimetry phantoms are to be used for assessing CTDIvol in relation to examinations on adults and children.
How should CTDIvol be determined when AEC is used?
Clarify the intended purpose of CTDIvol and DLP as being for comparison (as mean values for a group of patients) against reference doses set for typical CT examinations, as a practical aid to optimisation studies, rather than being indicators of dose (such as organ dose) for individual patients.
38. Section A2
Provide clearer guidance on whether DLP is still a valid quantity for MDCT and how the effects of AEC should be incorporated.
39. Section A3
Line 1834: Clarify that background radiation is typically in the range 1 to 3 mSv per year.
Table A1: The range of effective dose quoted for ‘Chest radiograph’ (0.1 - 0.2 mSv) is rather higher than typical practice in the UK (around 0.02 mSv) (Hart and Wall, UK population dose from medical x-ray examinations, Eur J Radiol 50 (2004), 285-291).
Table A2: Updated E/DLP coefficients for adult and paediatric patients have been published as part of the 2004 European quality criteria for MSCT (http://www.msct.eu/CT_Quality_Criteria.htm#Download%20the%202004%20CT%20Quality%20Criteria) and in the reference Shrimpton et al 2005 (and most recently in Shrimpton et al, National survey of doses from CT in the UK: 2003, Br J Radiol 79 (2006), 968-980).
Line 1866: The use of ‘reference values’ here could be confusing (given this term’s particular use in medical dosimetry).
Line 1871: Should effective dose be used to characterise dose (and risk) for therapeutic procedures (radiotherapy?), where deterministic effects are likely to arise from the irradiation?
40. Section A4
This presents a rather brief and selective overview of the dose coefficients available for CT. A more comprehensive review would be useful.
It should be mentioned that such dose coefficients generally require values of the dose (free-in-air) on the axis of rotation of the scanner in order to estimate typical patient doses.
Line 1885: Hart et al 1994 present dose coefficient in relation to conventional x-ray examinations on adult patients, rather than CT.
Line 1886: The reference Shrimpton et al 1991 should be replaced with Jones DG and Shrimpton PC, Normalised organ doses for x-ray computed tomography calculated using Monte Carlo techniques, Chilton, NRPB-SR250, 1993.