BJR
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
British Journal of Radiology (2003) 76, 798-802
© 2003 British Institute of Radiology
doi: 10.1259/bjr/33117342
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Koller, C J
Right arrow Articles by Bettridge, A
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Koller, C J
Right arrow Articles by Bettridge, A

Full Paper

Variations in radiation dose between the same model of multislice CT scanner at different hospitals

C J Koller, MSc 1 J P Eatough, PhD 1 and A Bettridge, DCR(R) 2

1 Medical Physics Directorate, University Hospital of North Staffordshire NHS Trust, Princes Road, Hartshill, Stoke on Trent ST4 7LN and 2 X-ray Department, Queens Hospital, Belvedere Road, Burton-on-Trent ST16 3SA, UK


   Abstract
Top
 Abstract
Introduction
 Method
 Results
 Discussion
 Conclusion
 References
 
The variation in exposure factors and patient dose, between seven centres using identical multislice CT scanners, was investigated for six standard examinations. Dose values were compared with each other and the relevant diagnostic reference level (DRL) for each examination. The range in weighted CT dose index (CTDIw) values between the seven centres was small for abdominal scans and head scans. For other scans however, such as functional endoscopic sinonasal surgery (FESS) the variations in CTDIw were as high as a factor of seven between the lowest and the highest values. At one centre a program of dose optimization had been undertaken and this centre had CTDIw values ranging from 3% to 64% lower than the average value for the seven centres. This demonstrates that significant dose reduction can be achieved through close collaboration between medical physicists, radiologists and radiographers.


   Introduction
Top
 Abstract
 Introduction
Method
 Results
 Discussion
 Conclusion
 References
 
It is well recognised that CT is a comparatively high dose diagnostic X-ray procedure. In 1989 the National Radiological Protection Board showed that despite comprising only 2% of all examinations, CT contributed around 20% of the collective dose to the UK population from diagnostic X-ray imaging [1]. The use of CT has been steadily increasing and more recent data indicate that CT now accounts for 40% of the collective dose to the UK population from medical X-rays [2, 3]. This increase may be due in part to an increase in the number of CT scanners available and also to the increasing sophistication of scanner technology such as spiral scanning, 3D reconstruction, real-time fluoroscopic CT, multislice scanning and sub-second scan times. All of these advances may have an impact on image quality, scan time and patient dose.

The Ionising Radiation (Medical Exposure) Regulations (IR(ME)R) 2000 require that image quality and radiation dose are optimized for every medical exposure. The relative importance of CT to patient dose in diagnostic radiology is recognised by the further requirement that "special attention" be paid to dose optimization for this practice [4]. However, dose optimization in CT is not straightforward. Variations in equipment design between manufacturers and models, such as the level of filtration, different focus–isocentre distances, and variations in collimator and detector efficiency can have a significant impact on dose [5]. Lack of any "automatic exposure control" on most existing CT scanners, combined with the wide dynamic range of digital detectors may allow wide differences in user-selectable factors between different CT units. Factors affecting patient dose include the choice of slice width, pitch, kV, mAs per rotation and the number of slices per examination.

European guidelines on CT imaging have been available for several years [6]. These guidelines detail the required image quality and suggest diagnostic reference levels (DRLs) for nine examinations. Of these reference levels only four have been adopted in the UK [7], and much of the data was collected from a UK survey undertaken in the early 1990s [1], prior to the widespread introduction of spiral scanning.

Previous studies of patient dose in CT have typically included a variety of examinations on a range of different scanners [8–10], and variations in patient dose in such surveys will be affected by both inherent equipment design and user-selectable factors. However, practical dose optimization for each exposure relates principally to the user-selectable exposure factors.

The purpose of this study was first to identify the variation in user-selectable exposure factors employed by different hospitals for each of a range of standard examination protocols carried out on the same model of multislice CT scanner; second to determine the effect of these variations on patient dose; and third to compare the range of patient doses with the relevant DRL values. By restricting this study to a single model of CT scanner, variations in patient dose owing to equipment design are removed. In addition each unit must meet the manufacturer's specifications for image quality, and this is verified independently during commissioning and routine surveys. Hence each CT scanner can achieve comparable image quality. This means that the variations in patient dose will be a direct result of the variation in user-selectable exposure factors.


   Method
Top
 Abstract
 Introduction
 Method
Results
 Discussion
 Conclusion
 References
 
A questionnaire was sent to 12 hospitals that had installed a Toshiba (Toshiba Medical Systems, NL) Asteion multislice CT scanner within the previous 12 months. Standard scanning protocols were requested for the following examinations: abdominal scan, chest scan, high-resolution chest scan and view (HRCT S & V), neck soft tissue scan, head scan and sinus scan for functional endoscopic sinonasal surgery (FESS).

Dose measurements were made on one Toshiba Asteion Multislice CT using a 10 cm long CT pencil ionization chamber (Victoreen, USA) positioned free in air along the central axis of the scanner. Measurements were made for a range of different slice widths and exposure factors. The CT dose index, CTDI10cm,air [11] for each of these measurements was then calculated: Go


where R is the chamber reading in mGy, cfe is the chamber calibration factor, L is the chamber length (cm), N is the number of simultaneous slices and T is the nominal slice thickness (cm).

This CT chamber was also used to make dose measurements within a 16 cm head and 32 cm diameter body polymethyl methacrylate standard CT dosimetry phantom as specified by the Food and Drug Administration [12]. The weighted CT dose index, CTDIw [6] was then calculated for each phantom using typical exposure factors: Go


where CTDI10cm is calculated in the same manner as CTDI10cm,air described above (Equation 1Go), CTDI10cm,centre is measured at the centre of the phantom and CTDI10cm,periphery is measured at the periphery of the phantom.

Using the results from the questionnaires and by applying scaling factors (determined from the CTDI10cm,air measurements) to the measurements made above, CTDIw values were calculated for all of the CT scans involved in the study.

It is anticipated that there will be variations in dose between the different CT scanners, resulting from differences in X-ray beam collimation, generator factors such as kVp, ripple and mAs calibration, and X-ray tube target angle. The manufacturers specify a ±20% margin for variations in CTDIw between scanners (personal communication: Toshiba, 2002), but it is understood that this includes an allowance for the positional errors resulting from variation in measurement technique. A more accurate method to compare tube outputs is the measurement of CTDIair, as this is measured at the isocentre and is less prone to systematic positioning errors and the effects of over-scan for example. Data from a national survey undertaken by ImPACT (St Georges Hospital, London) in 1996 give a variation in CTDIair of 7% between identical models, however this also includes "operator" dependent variables such as differences in equipment and chamber calibrations. Given that our results are based on CTDIair measurements the variation in CTDIair between multiple scanners of the same model is likely to be of the order of 5% (personal communication: ImPACT, 2002).


   Results
Top
 Abstract
 Introduction
 Method
 Results
Discussion
 Conclusion
 References
 
Of the 12 hospitals contacted, seven replies were received and the results are summarized in Table 1Go. As expected the range in kV and scan time between hospitals for a given examination was small. There was a larger variation in the selected slice width, however the greatest variation was in the mA selected for a particular examination. In one case, the FESS scan, the largest selected mA was four times the smallest.


View this table:
[in this window]
[in a new window]
  		Table 1. Summary of survey questions and responses received from seven scanners at different Trusts in the UK

 
At one hospital (E) an active programme of dose and image quality optimization had been implemented. This had involved gradual adjustment of exposure factors in consultation with medical physics staff, whilst the quality of the images obtained was assessed by a consultant radiologist to ensure they were clinically satisfactory, as well as met the relevant DRLs (Table 2)Go and the required European guidelines [6].


View this table:
[in this window]
[in a new window]
  		Table 2. Source of CTDIw reference values marked on the graphs in figure 1Go

 
For abdominal examinations the CTDIw values were well below the DRL and showed little variation between most of the hospitals (Figure 1aGo). One hospital (E) achieved a CTDIw value 32% lower than the other hospitals following careful choice of slice thickness and tube current. In addition the questionnaire requested exposure parameters for both average and large patients. It is interesting to note that of the seven hospitals that replied, only two of these hospitals set different parameters for average and large patients.



View larger version (27K):
[in this window]
[in a new window]
  		Figure 1. Weighted CT dose index (CTDIw) values for: (a) routine abdominal scans; (b) routine head examinations; (c) chest high resolution CT (HRCT) scans; (d) chest scans (non-HRCT); (e) functional endoscopic sinonasal surgery (FESS) scans; and (f) neck soft-tissue scans. For sources of reference value data see Table 2.

 
The CTDIw values for the posterior fossa area for head examinations exceeded the reference level for routine head examinations (60 mGy) at all hospitals (Figure 1bGo). The lowest value was 19% less than the mean value for all the hospitals. For cerebral scans, the CTDIw values were marginally less than the reference level for five of the seven hospitals, and a smaller variation was found between hospitals.

For chest high resolution CT (HRCT) scan there was a factor of more than two between the highest and lowest CTDIw values (Figure 1cGo). The choice of slice width varied from 0.5 mm to 1.0 mm between hospitals. For standard chest scans (non HRCT) (Figure 1dGo), there was a smaller variation between the highest and lowest CTDIw values.

An even larger variation was observed for FESS scans (Figure 1eGo), where there was a factor of seven between the highest and lowest CTDIw values.

For the neck soft-tissue scans (Figure 1fGo) the CTDIw values varied little between six of the hospitals. At one hospital (G), however, the calculated CTDIw value was around a quarter of the mean value for all the hospitals.

Overall, hospital (E) had the lowest CTDIw values for four of the six examinations. The CTDIw values were 24% less than the average value for abdominal scans; 9% less overall for routine head scans; 47% less than the average for HRCT chest scans; 7% less than the average for non-HRCT chest scans; 64% less than the average for FESS and 3% less for neck soft-tissue scans.


   Discussion
Top
 Abstract
 Introduction
 Method
 Results
 Discussion
Conclusion
 References
 
It is known that large variations in CT practice exist [13], particularly in the choice of user-selectable exposure parameters, scanned area, number of slices and pre/post contrast views. The introduction of a new CT scanner is a good opportunity to review local practices against national recommendations. This is particularly true with the introduction of multislice scanning, which is relatively new technology. In this instance users initially look to the manufacturer for guidance on suitable exposure factors for different examinations. When experience has been gained on the equipment, users can then begin to optimize exposure parameters.

In all seven hospitals, the reference level for abdominal scans was easily achieved. This may imply that either the scanning parameters have been optimized, or that the DRL is now too high for modern scanners because it is based on CT data collected in the early 1990s when spiral scanning was in its infancy.

More interesting however is that only two hospitals adjust their exposure parameters according to patient size. Larger patients will introduce more photon scatter and attenuation, and if the scanning parameters are not altered the noise present in the images for large patients will be greater than that for average patients. Dose reductions of up to 45% can be achieved by adjusting exposure factors with patient size [14], which indicates that although the abdominal scans meet the DRL there still exists further room for optimization.

For chest scans (non HRCT) the CTDIw values followed a similar pattern to abdominal scans. All hospitals were lower than the 30 mGy DRL, but by comparing these values with the results obtained at centre E, there still exists scope in the other hospitals for dose reduction.

The CTDIw values for routine head scans were close to, or exceeded the DRL. For the cerebral region five hospitals achieved a CTDIw lower than the reference value, however the average value for all seven hospitals exceeded the DRL. For the posterior fossa region all hospitals exceeded the DRL. This is not unusual as other studies indicate that this is a difficult target to achieve [9]. Scans through this region of the brain utilize narrow slices (typically 2 mm). These high dose values on multislice CT scanners may well be due to engineering design and there may be little scope for further reduction. Such scanners have a lower geometric efficiency compared with single slice CT scanners. This is particularly apparent for narrow slice widths where the geometric efficiency can fall to below 50% [15]. Hence half of the radiation dose received by the patient does not contribute to the image.

Both the HRCT and FESS scans showed a large variation in CTDIw values between hospitals. It is interesting to note the similarities between these examinations. Both contain areas of the body that generate high contrast images, i.e. they contain air, tissue and bone, making good quality images relatively easy to achieve. In addition both scans involve the identification of relatively small structures within the body. This requires good spatial resolution, achieved through the selection of a narrow slice width (typically 1–2 mm).

The dose-limiting factor with such narrow slice selections is the noise present in the image. For narrow slice widths the number of X-ray photons received at the detector is reduced, compared with a wider slice width under similar conditions of exposure. In many cases the mAs per scan is increased to maintain the same signal to noise ratio in the image. However, where high contrast details are being visualized, the presence of increased noise may not be deleterious to the diagnosis. This is an area that requires further investigation to ensure that images are of sufficient quality to answer the diagnostic question but patient doses are kept as low as reasonably practicable.

The variation in CT dose found here is consistent with findings from other studies [8–10]. It is a requirement of IR(ME)R that "special attention" be paid to the optimization of high dose procedures and this requires a multidisciplinary approach. This paper has highlighted the significance of the choice of user-selectable exposure parameters, and also the dose savings that can be achieved through close collaboration between medical physicists, radiographers and radiologists.


   Conclusion
Top
 Abstract
 Introduction
 Method
 Results
 Discussion
 Conclusion
References
 
This survey has shown a considerable variation in the choice of user-selectable exposure factors for some common types of examination, carried out on the same model of CT scanner, at a number of different hospitals. The resulting variation in CT dose was due to the variation in user-selectable exposure factors alone.

For abdominal scans the DRL was achieved and the variation between hospitals was small. However, there still remains scope for further reduction of patient doses, particularly when due consideration is given to patient size while selecting exposure factors.

For HRCT and FESS examinations there was a considerable variation between hospitals, implying that there is substantial scope for dose optimization.

At one hospital, where dose optimization has been implemented, dose savings between 3% and 64% have been achieved whilst maintaining diagnostic image quality. This has only been achieved through close collaboration between radiographers, physicists and radiologists.


   Acknowledgments
 
The authors wish to thank the Radiologists at Queen's Hospital Burton, who assessed numerous CT images to ensure that they met the diagnostic requirements, as identified by the European Guidelines. Thanks are also due to the radiographic staff at the other hospitals for participating in this study.

Received for publication October 22, 2002. Revision received May 30, 2003. Accepted for publication June 10, 2003.


   References
Top
 Abstract
 Introduction
 Method
 Results
 Discussion
 Conclusion
 References
 

  1. Shrimpton PC, Jones DG, Hillier MC, Wall BF, Le Heron JC, Faulkner K. Survey of CT practice in the UK. Part 2: dosimetric aspects. NRPB R249. Chilton: NRPB, 1991.
  2. Mettler FA Jr, Wiest PW, Locken JA, Kelsey CA. CT scanning: patterns of use and dose. J Radiol Prot 2000;20:353–9.[CrossRef][Medline]
  3. Shrimpton PC, Edyvean S. CT scanner dosimetry. Br J Radiol 1998;71:1–3.[Medline]
  4. Ionising Radiation (Medical Exposure) Regulations 2000 (Statutory Instrument No. 1059) (together with notes for good practice). London: The Stationary Office, 2000.
  5. Dixon AK, Dendy P. Spiral CT: how much does radiation dose matter? The Lancet 1998;352:1082–3.[CrossRef][Medline]
  6. European Guidelines on Quality Criteria for Computed Tomography, EUR 16262 EN. Luxembourg: Office for Official Publications of the European Committees, 1999.
  7. Guidelines on patient dose to promote optimisation of protection for diagnostic medical exposures. Documents of the NRPB 1999, vol 10, No 1. Chilton: NRPB.
  8. Clarke J, Cranley K, Robinson J, Smith PHS, Workman A. Application of draft European Commission reference levels to a regional CT dose survey. Br J Radiol 2000;73:43–50.[Abstract]
  9. White MJ, Eatough JP, Goldstone KE. Audit of effective dose for patients undergoing X-ray Computed Tomography in the East Anglian Region. Report produced by East Anglian Regional Radiation Protection Service, 1999.
  10. Hiles PA, Brennen SE, Scott SA, Davies JH. A survey of patient dose and image quality for computed tomography scanners in Wales. J Radiol Prot 2001;21:345–54.[Medline]
  11. Edyvean S. Type testing of CT scanners: methods and methodology for assessing imaging performance and dosimetry. MDA/98/25 London: HMSO, 1998.
  12. Food and Drug Administration, department of health and human services. Computed Tomography Equipment. Performance standard for diagnostic x-ray systems. FDA US Government Printing Office, 1988
  13. Golding SJ, Shrimpton PC. Radiation dose in CT: are we meeting the challenge? Br J Radiol 2002;75:1–4.[Free Full Text]
  14. Starck G, Lonn L, Cederbald A, Forssell-Aronsson E, Sjostrom L, Alpsten M. A method to obtain the same levels of CT image noise for patients of various sizes, to minimise radiation dose. Br J Radiol 2002;75:140–50.[Abstract/Free Full Text]
  15. ImPACT four slice CT scanner comparison, MDA 02061. London: HMSO, 2002



This article has been cited by other articles:


Home page
 Am. J. Roentgenol.Home page
T. A. Jaffe, T. T. Yoshizumi, G. Toncheva, C. Anderson-Evans, C. Lowry, C. M. Miller, R. C. Nelson, and C. E. Ravin
Radiation Dose for Body CT Protocols: Variability of Scanners at One Institution
Am. J. Roentgenol., October 1, 2009; 193(4): 1141 - 1147.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Neuroradiol.Home page
C.B. Nauer, A. Eichenberger, P. Dubach, J. Gralla, and M. Caversaccio
CT Radiation Dose for Computer-Assisted Endoscopic Sinus Surgery: Dose Survey and Determination of Dose-Reduction Limits
AJNR Am. J. Neuroradiol., March 1, 2009; 30(3): 617 - 622.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Roentgenol.Home page
T. Kubo, P.-J. P. Lin, W. Stiller, M. Takahashi, H.-U. Kauczor, Y. Ohno, and H. Hatabu
Radiation Dose Reduction in Chest CT: A Review
Am. J. Roentgenol., February 1, 2008; 190(2): 335 - 343.
[Abstract] [Full Text] [PDF]

Home page
 Br. J. Radiol.Home page
A J Van Der Molen, W J H Veldkamp, and J Geleijns
16-slice CT: achievable effective doses of common protocols in comparison with recent CT dose surveys
Br. J. Radiol., April 1, 2007; 80(952): 248 - 255.
[Abstract] [Full Text] [PDF]

Home page
 RadiologyHome page
V. Tsapaki, J. E. Aldrich, R. Sharma, M. A. Staniszewska, A. Krisanachinda, M. Rehani, A. Hufton, C. Triantopoulou, P. N. Maniatis, J. Papailiou, et al.
Dose Reduction in CT while Maintaining Diagnostic Confidence: Diagnostic Reference Levels at Routine Head, Chest, and Abdominal CT--IAEA-coordinated Research Project
Radiology, September 1, 2006; 240(3): 828 - 834.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Koller, C J
Right arrow Articles by Bettridge, A
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Koller, C J
Right arrow Articles by Bettridge, A

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
BJR DMFR IMAGING  ALL BIR JOURNALS