Intra-institution CT dose surveys in adults: what sample size for what precision?

Size: px

Start display at page:

Download "Intra-institution CT dose surveys in adults: what sample size for what precision?"

Cornelia Kelley
7 years ago
Views:

1 Intra-institution CT dose surveys in adults: what sample size for what precision? Poster No.: B-0785 Congress: ECR 2015 Type: Authors: Scientific Paper S. Taylor 1, A. Van Muylem 2, P. A. Gevenois 2, D. Tack 3 ; 1 Mons/BE, 2 Brussels/BE, 3 Baudour/BE Keywords: DOI: Radioprotection / Radiation dose, CT, Dosimetry /ecr2015/B-0785 Any information contained in this pdf file is automatically generated from digital material submitted to EPOS by third parties in the form of scientific presentations. References to any names, marks, products, or services of third parties or hypertext links to thirdparty sites or information are provided solely as a convenience to you and do not in any way constitute or imply ECR's endorsement, sponsorship or recommendation of the third party, information, product or service. ECR is not responsible for the content of these pages and does not make any representations regarding the content or accuracy of material in this file. As per copyright regulations, any unauthorised use of the material or parts thereof as well as commercial reproduction or multiple distribution by any traditional or electronically based reproduction/publication method ist strictly prohibited. You agree to defend, indemnify, and hold ECR harmless from and against any and all claims, damages, costs, and expenses, including attorneys' fees, arising from or related to your use of these pages. Please note: Links to movies, ppt slideshows and any other multimedia files are not available in the pdf version of presentations. Page 1 of 11

2 Purpose Basing our work on the current dose sruvey recommendations applied within the EU (1-9), our aims were: To determine the variability of dose-length product (DLP) per CT acquisition; To determine the minimum sample size to warrant sufficient reliability. Methods and materials #Patients #Anonymized data of 10,663 CT examinations performed between January and December 2013 (Head: 2,365; #Abdomen: 2,237; #Thorax: 2,026; #Lumbar spine: 2,011; #Cervical spine: 1,237; #Sinuses: 787); All examinations were conducted on two identical commercially available 64 detector-rows CT devices providing 128 slices per rotation (Definition AS+, Siemens Healthcare, Forchheim, Germany) with strictly the same acquisition parameters and using automatic exposure control (AEC) (CARE Dose4D, Siemens Healthcare); #Patients' weight recorded for thoracic, abdominal, and lumbar spine examinations; #When body weight was considered, three categories were considered: #No weight selection; Kg; #60-80 Kg. #Statistical analysis #All distributions tested for normality of their distribution (Lilliefors variant of the Kolmogorov-Smirnov test): All differ significantly from a normal distribution (P<.001); 2,000 samples drawn randomly from each population of DLP values, for given organ and weight category; The calculated dispersion of each distribution was based on the 95% confidence interval (CI95) in percentage of the median (med); The dispersion was calculated for increasing sample sizes; We calculated the sample size that set a 95% CI lower than 10% of the median (CI95/med<10%). Results Page 2 of 11

3 Figure 1-6 show the results per body region with no body selection and increasing sample sizes ( Fig. 1 on page 8, Fig. 2 on page 7, Fig. 3 on page 3, Fig. 4 on page 4, Fig. 5 on page 5, Fig. 6 on page 6 ). Figure 7 shows the required sample sizes to reach CI95/med<10% for each body region and in relation to body weight selection. Overall, our results show that: 1. DLP delivered by CT with AEC varies greatly between individuals; 2. The smaller the sample, the greater the variability of mean DLP; 3. The sample size required for ensuring CI95/med<10% depends on the body region; 4. Even with a body weight selection, CI95/med<10% is not reached with samples as large as 50 patients scanned at the level of the trunk; 5. None of the data collection methods currently recommended by regulatory authorities for CT dose surveys reaches CI95/med<10%; 6. Substantial differences in median or mean DLP values of a given type of examination are associated with the body weight selection 7. With a sample size of 10 patients, the CI95/med<10% of a local dose descriptor, the latter being only 30% higher or lower than the DRL, would include the DRL itself. This can have two consequences. If this local dose descriptor is above the DRL but its true value is bellow the DRL, it can be inaccurately considered as a need of optimization. Likewise, if the local dose descriptor is below the DRL but its true value is above the DRL, it can be inaccurately considered to not be in need of optimization. Our study had limitations: 1. The data was collected from a single CT system from one manufacturer with its particular AEC system. 2. As AEC system was switched on for all our acquisitions, our results could not be generalized to radiology departments with other practices. 3. We arbitrarily aimed to achieve a variability of CI95/median<10%, whereas one could aim to reach an even lower variability. 4. We did not address the variability of multiphasic CT examinations, and did not take into account their clinical indications. Images for this section: Page 3 of 11

4 Fig. 3: Cervical spine - Shows the 95% confidence interval (open circles) in percentage of the median as a function of the sample size. Vertical line corresponds to the sample size ensuring CI95/med<10%. Page 4 of 11

5 Fig. 4: Lumbar spine - Shows the 95% confidence interval (open circles) in percentage of the median as a function of the sample size. Vertical line corresponds to the sample size ensuring CI95/med<10%. Page 5 of 11

6 Fig. 5: Head - Shows the 95% confidence interval (open circles) in percentage of the median as a function of the sample size. Vertical lines correspond to the sample size ensuring CI95/med<10%. Page 6 of 11

7 Fig. 6: Sinus - Shows the 95% confidence interval (open circles) in percentage of the median as a function of the sample size. Vertical lines correspond to the sample size ensuring CI95/med<10%. Fig. 7: Sample size required to reach a CI95<10% Page 7 of 11

8 Fig. 2: Abdomen - Shows the 95% confidence interval (open circles) in percentage of the median as a function of the sample size. Vertical line corresponds to the sample size ensuring CI95/med<10%. Page 8 of 11

9 Fig. 1: Thorax - Shows the 95% confidence interval (open circles) in percentage of the median as a function of the sample size. Vertical line corresponds to the sample size ensuring CI95/med<10%. Page 9 of 11

10 Conclusion Variability of DLP in actual CT radiation dose surveys is high using data samples of ten, with the risk of including the DRLs within the range, therefore implying that reducing the variability impacts on the sample size with a four to ninety fold increase according to body region scanned. Personal information References 1. Hricak H, Brenner DJ, Adelstein SJ, Frush DP, Hall EJ, Howell RW, et al. Managing Radiation Use in Medical Imaging: A Multifaceted Challenge. Radiology 2011; 258; Shrimpton PC, Wall BF, Jones DG, Fisher ES, Hillier MC, Kendall GM, Harrison RM. Doses to patients from routine diagnostic X-ray examinations in England. Br J Radiol 1986; 59: The council of the European Union (2013) Council Directive 2013/59/ Euratom of 5 December 2013 on Laying down basic safety standards for protection against the dangers arising from exposure to ionising radiation, and repealing Directives 89/618/Euratom, 90/641/Euratom, 96/29/Euratom, 97/43/Euratom and 2003/122/Euratom. Access on line on January 25, 2015 at: uri=oj:l:2014:013:0001:0073:en:pdf 4. European commission (1998) Radiation Protection 102: Implementation of the medical exposure Directive (97/43/Euratom). Proceedings of the international workshop held in Madrid on 27 April Access on line on January 25, 2015 at: documents/102_en_0.pdf 5. European commission (1999) Radiation Protection 109: guidance on diagnostic reference levels for medical exposures. EC1999 Access on line on January 25, 2015 at: documents/109_en_0.pdf 6. Shrimpton PC, Hillier MC, Lewis MA, Dunn M National survey of doses from CT in the UK: Br J Radiol.;79: Shrimpton PC, Jessen KA, Geleijns J, Panzer W, Tosi G (1998) Reference doses in computed tomography. Radiat Prot Dosim;80: Stamm G. Collective radiation dose from MDCT. Critical review of survey studies. In TackD, Kalra MK, Gevenois PA (eds). Radiation dose from multidetector CT Springer Heidelberg. pp Page 10 of 11

11 9. Tack, D., Jahnen, A., Kohler, S., Harpes, N., De Maertelaer, V., Back, C., & Gevenois, P. A. (2013). Multidetector CT radiation dose optimisation in adults: short- and long-term effects of a clinical audit. Eur Radiol 2014;24: Page 11 of 11

CT findings in Differential Diagnosis between Tuberculous Pleurisy and Malignant Effusion

CT findings in Differential Diagnosis between Tuberculous Pleurisy and Malignant Effusion Poster No.: E-0084 Congress: ESTI 2012 Type: Scientific Exhibit Authors: S. S. Shim, Y. Kim; Seoul/KR Keywords: