Standpoint of the Swiss authorities of protection against radiation General comments 1. The project approaches a burning topic of radiological protection in the medical field. The fast evolution of the rate of CT examinations and of the dose per examination was also observed in Switzerland. Thus between 1998 and 2003 the collective dose due to CT examinations increased by a factor 1,7. This increase is related on the one hand to the increase in the number of examination (factor: 1,4) and with the increase in the average dose per examination (factor: 1,2). 2. The step followed, that consist in not giving ready receipts, but to explain the mechanisms which influence the exposure of the patient and the quality of the image in scanning, is considered to be very relevant. Indeed the interactions between the parameters of adjustment and the quality of image are multiple and complicated. Only a detailed comprehension of operation allows a rational optimization of the procedures. 3. Within this framework it would be desirable to hold the operation of CT installations exclusively to the radiologists, because they are currently the only ones to have the base of knowledge necessary to exploit CT in a sure and optimal way. It would be also judicious to insist on the role of the medical physicist like partner with whole share in the step of optimization. 4. The dosimetric quantities selected should systematically correspond to those proposed by recent publication 74 of the international Commission on radiation units and measurements (ICRU). In particular the CTDI (line 1733) is not the absorbed dose, but a Kerma in air measured in a phantom of PMMA of clearly definite size. The use of the absorbed dose (lines 1734 and 1782) would require moreover to defining which material is concerned (here of the air, according to the recommended factor f). The fact of shortening the notation while speaking about CT dose index is certainly acceptable, insofar as this abuse language were indicated and justified. 5. In the examples of optimization of CT examinations it would be judicious to indicate the 3 principal dosimetric parameters systematically, namely CTDIvol, the PDL and the effective dose. In many tables (4.1; 4.2; 4.3; 4.4;4.5) information are incomplete. In particular it is often restricted to the indication of the effective dose. Particular comments 1. It is indicated (lines 227) that the MDCT uses a geometry of 3rd generation, which is correct. On the other hand in line 282 one speaks about 4th generation what is certainly incorrect. 2. The effective mAs is described (line 348) as the average charge (in mAs) per unit of length along the longitudinal axis. This is certainly incorrect; indeed its unit would be then the mAs/cm. Actually the effective mAs is obtained by dividing the mAs by slice per pitch (without unit) and is thus expressed in mAs. It is the mAs which would lead to the same average exposure of the patient for a pitch of 1. 3. Line 390 opens the door to CT exploitation by cardiologists. This appears to us in contradiction with the vision of a complex technology requiring of thorough knowledge of the radiological physics. 4. With the line 1232 one finds: "in MDCT an increase in pitch is associated with an increase in image noise". This is presented in opposition to the situation of the SDCT. It would be judicious to justify this assertion which at first view appears incorrect. 5. It would be necessary to replace "avoided" by "used" in line 1240, because it is more judicious to use only one acquisition in spiral than to break up it, this because of the overranging and the stepping of acquisitions. 6. In table 4.1.B a 2-4 folds dose reduction is mentioned in the legend, whereas in the table the maximum reduction is of 25 %. This contradiction is mainly due to the fact that the statements are different in the text and in the table. 7. In tables 4.2.B, 4.3.C and 4.5.B, a dose reduction is presented without indicating the standard procedure before optimization. This reduction could be rebuilt for the scanner in question and be indicated to facilitate the comparison. 8. In table 4.2.C a CTDIvol reduction of a factor 3 can be observed (reduction related on the effective mAs and on ECG gating). On the other hand the dose is reduced by a factor 2 whereas all the other conditions, including the length of the scan, remain the same ones. It would be judicious to explain this point. 9. The line 1699 promotes to use a "wider beam collimation" to protect the fetus. This methodology should be clarified. 10. In the line the 1834 the range of dose relates to the annual dose (1 to 3 mSv/an). Comments of detail 1. It would be useful that the points of the summary and those which introduce the chapters are the same ones. 2. Reference IRCU 2006 (line 424) does not appear in the bibliography. 3. The sentence of lines 522 to 523 is broken. 4. The reference of line 1041 (ICRP, 1991), i.e. publication ICRP 60, appears inadequate. 5. In table 4.4.B it would be judicious to indicate the noise index in the first column. Additional comments from a practical point of view. A long text from over 80 sides risked by the practitioner at the front not to be read; to solve this problem the current summary (Summary POINTS) could be supported by an illustration/table showing connections and overview. 1. Paragraph 3.7. (line 1243) about indication appears to briefly in view of the clinical meaning. 2. Line 1559 and the following, tab. 4.3.A-C: CT Colonography: according to available literature ~ 2-3 mSv seems to be an acceptable dose, because the presence of the liquid requires to scan twice (in moving and prone position, supine and prone); Tab. 4.3.A appears untenable deeply, while one must really ask oneself whether there are really indications for children CT Colonographie (tab. 4.3.C) 3. Line 1648, table 4.4.B: why working here with 140 kV? All three protocols lead to very high doses for the renal stones diagnostics, while against it for the same question in tab. 4.4.A an unrealistic "low dose protocol" is introduced, that means less than 10% of the deepest effective dose given in 4.4.B. An intermediate way would seem here more realistic 4. Line 1716 tab. 4.6. The effective doses for all three protocols seem - even without pregnancy - very high, again partly due to the used 140 kV instead of e.g. 120 kV. In a pregnancy it should be possible to drive pulmonary embolism protocols with effective doses clearly under 10 mSv, appendicitis clarification also under 10 mSv and renal stones clarification e.g. with 2-3 mSv. Thus also the fetal doses - dependent on the age of the pregnancy - would sink.