Radiology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published online before print March 13, 2007, 10.1148/radiol.2432050930
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
2432050930v1
243/2/461    most recent
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
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 Vaño, E.
Right arrow Articles by de Las Heras, H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Vaño, E.
Right arrow Articles by de Las Heras, H.
(Radiology 2007;243:461-466.)
© RSNA, 2007

Medical Physics

Transition from Screen-Film to Digital Radiography: Evolution of Patient Radiation Doses at Projection Radiography1

Eliseo Vaño, PhD, José Miguel Fernández, MSc, José Ignacio Ten, MSc, Carlos Prieto, MSc, Luciano González, PhD, Ricardo Rodríguez, MD, PhD, and Hugo de Las Heras, MSc

1 From the Medical Physics Service (E.V., J.M.F., C.P., H.d.L.H.) and Radiology Service (J.I.T., R.R.), San Carlos University Hospital, 28040 Madrid, Spain; and the Radiology Department, Complutense University, Madrid, Spain (E.V., J.M.F., L.G., R.R.). Received June 3, 2005; revision requested July 29; revision received June 9, 2006; accepted July 19; final version accepted September 8. Supported in part by the European Commission (DIMOND and SENTINEL programs), the Spanish Ministry for Science and Technology (project BFI2003-09434), the Spanish Ministry of Health (Directorate General of Public Health), and the Autonomous Community of Madrid (project GR/SAL/0272/2004). Address correspondence to L.G. (e-mail: luciano{at}med.ucm.es).


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 ADVANCES IN KNOWLEDGE
 References
 
Purpose: To retrospectively evaluate patient radiation doses in projection radiography after the transition to computed radiography (CR) in the authors' hospital.

Materials and Methods: The hospital's ethical committee approved the study and waived informed consent. In 2001, a dose reduction initiative was implemented, which involved collecting radiographic parameters, calculating patient entrance doses, and monitoring changes with an online computer, and a training program for radiographers was conducted. A database with 204 660 patient dose values was used to compute changes in patient doses over time. Sample sizes ranged from 1800 to 23 000 examinations. Doses were compared with European and American reference values. Kruskal-Wallis and Mann-Whitney tests were used for statistical analysis.

Results: Median values for patient entrance doses increased 40%–103% after implementation of CR. Initial increases were corrected during the 1st year, and additional dose decreases were achieved after the dose reduction initiative was launched. At present, doses range between 15% and 38% of the European diagnostic reference levels established for screen-film radiography and between 28% and 41% of the reference values recommended by the American Association of Physicists in Medicine, representing an effective 20%–50% reduction in the initial values for CR.

Conclusion: Though patient doses can increase considerably during the transition from conventional screen-film radiography to CR, dose management programs, including specific training of radiographers and patient dose audits, allow for reductions of the previous values.

© RSNA, 2007


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 ADVANCES IN KNOWLEDGE
 References
 
The transition from conventional screen-film to computed or digital radiography can entail an increase in patient radiation doses (1). One of the main causes for the increase is the wide dynamic range of the digital imaging systems, which allows overexposure with no adverse effect on image quality. In addition, the lack of specific training in the new digital techniques for some radiographers and the lack of well-established methods to audit patient doses in digital systems can worsen the problem of patient radiation exposure.

The International Commission on Radiological Protection (ICRP) became aware of this risk and launched several specific recommendations to manage patient doses in digital and computed radiology (1). These recommendations include appropriate training, particularly in aspects of patient dose management, revision of the diagnostic reference levels, and frequent patient dose audits. In addition, the ICRP recommended that the industry promote tools to inform radiologists, radiographers, and medical physicists about exposure parameters and the resultant patient doses.

Peters and Brennan (2) provided a warning about the likelihood of initial high doses with the use of settings recommended by manufacturers of computed radiography (CR) systems. In a retrospective analysis of 717 exposures at mobile chest examinations, they showed a substantial reduction in exposure parameters from the initial values and emphasized the need for clinicians and personnel to optimize procedures when moving to digital techniques.

In a randomized controlled trial, Weatherburn et al (3) compared the radiation doses received by patients during bedside chest radiography with those at a CR system and with those at a 400-speed screen-film system, demonstrating that the entrance surface doses were 31% higher in the CR group. The purpose of our study was to retrospectively evaluate patient radiation doses in projection radiography after the transition to CR in our hospital.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 ADVANCES IN KNOWLEDGE
 References
 
Imaging Devices
For the examination types monitored in this work, from 1999 to 2001, patients were imaged in one of three dedicated rooms, equipped with Optimus 50 generators (Philips Medical Systems, Best, the Netherlands) and devoted to general projection radiography, in a university hospital with 965 beds and 336 840 radiologic examinations in 2004. The rooms are linked to a personal computer through patient data organizer systems (Philips). All three radiographic systems have automatic exposure control devices, properly adjusted by the technical service of the manufacturer for screen-film or equivalent CR systems of nominal speed class 400, and are submitted to periodic quality control (QC) by the medical physics service of the hospital. Measured half-value layer values at 80 kVp were between 3.9 and 4 mm Al (typical values for these units). The same rooms, equipped with other radiography units, were used through 1997 and 1998 for conventional screen-film radiography, also adjusted to a speed class of 400.

Since 1999, photostimulable phosphor plates (models MD10, MD30, and MD40; Agfa-Gevaert, Mortsel, Belgium) have been used for digital imaging with several CR systems (ADC Compact; Agfa-Gevaert) and their corresponding workstations. Though improvements in image quality may have been noted when plates MD30 or MD40 are used instead of model MD10, exposure settings were not changed, because no specific recommendation was given by the manufacturer. Moreover, older and newer plates were used during the same time period for every room and every examination type.

For each type of radiographic examination, the imaging parameters (kilovolt peak, focal spot size, automatic exposure control chamber) archived in the radiography system and set in accordance with the European guidelines on quality criteria for diagnostic radiography (4), are automatically selected for a source-to-skin distance (SSD) also specified in the guidelines. The radiographer in charge may sometimes change the radiographic technique by using the manual mode instead of automatic exposure control and choosing different SSD, kilovolt peak, and milliampere-second settings according to individual patient characteristics or the radiographer's own preferences.

A computer program based on Visual Basic (Microsoft, Redmond, Wash), designed to perform online dose monitoring and provide radiographic technique details, retrieves tube milliampere-second, kilovolt peak, field size, and source-to-detector distance from each patient data organizer and calculates the entrance surface dose by using the x-ray tube output, which is measured periodically as part of the QC program. (The entrance surface dose, or patient entrance dose, is the absorbed dose in air at the surface of the patient in the center of the irradiated area, including the backscattered radiation from the patient.) For each examination type, a standard patient thickness is assumed for entrance dose estimation. The computer application also allows online comparison of the mean patient dose value for a recent sample with the local diagnostic reference levels in order to audit dose levels and introduce corrective action if necessary (5).

Doses monitored with the described system were obtained with undercouch or onwall stand Bucky units (Philips). In both cases, the SSDs were measured with a sensor linked to the position of the x-ray tube and reported along with the radiographic technique data. The backscatter factor was assumed to be 1.35 for all examinations, as recommended by the European guidelines document (4). Examinations obtained without Bucky units were not analyzed, because SSDs are not available from the patient data organizer in these cases.

Until 2001, the operation of the new digital systems installed in 1999 was demonstrated by the vendors in a few short information sessions. No specific training to optimize patient dose or image quality was initially included. In 2001, a training program for radiographers and radiologists was provided by the hospital's medical physics service. It was conducted over 10 hours during 1 week, and its topics included general radiation protection concepts, introduction to digital imaging, CR and flat-panel radiography, recommendations for use, and patient dose management in CR. As manual exposure techniques were sometimes used, special attention was paid to the effects of changing the technical parameters on patient dose and the optimal kilovolt peak, milliampere-second, and SSD settings for every examination type. In addition, the QC computer application was launched as an automatic routine tool for online patient dosimetry auditing and was formally integrated into the hospital QC program.

Radiologists working in the monitored rooms evaluated the quality of clinical images yearly as part of the hospital QC program, by using random samples and the anatomic criteria recommended in the European guidelines (4). According to the QC reports, images had sufficient diagnostic quality during the years reported.

Patient Data
Our ethical committee approved our study and waived informed consent. For 1997–1999 for the screen-film systems, patient dose results are scarce, because radiographic data and distances were recorded manually only as required for the QC program. At this time, a less demanding local regulation required a sample of only 10 standard-sized patients per examination type per year, and the chest and lumbar spine were chosen as target examinations for QC.

Digital imaging was put into service in 1999. Since then, relevant procedures with diagnostic reference levels published in the European guidelines (4) (ie, chest, lumbar spinal, abdominal, and pelvic examinations), as well as thoracic spinal examinations (1800–23 000 examinations per examination type), were analyzed from a sample of 204 660 adult patients in the database who were examined in the three radiography rooms.

Statistical Analysis
Because patient dose distributions are skewed—especially in digital radiology, in which high doses do not detract from image quality—the nonparametric Kruskal-Wallis test was used for analysis. After that, the Mann-Whitney test was used to identify statistically significant differences (6) in annual dose values, comparing pairs of median dose values for each examination type from consecutive years. Instead of using the usual cutoff P value of .05, we assumed statistical significance for differences with P values less than .025 to account for the fact that each year's data were included twice in the comparisons. Software (SPSS, version 12.0, 2003; SPSS, Chicago, Ill) (7) was used for the tests. Because data before 1999 were scarce, a similar statistical analysis of that period has not been possible.

Because large samples were used (between 1800 and 23 000 examinations), median and interquartile values are the best statistical descriptors. However, for comparison purposes, the mean values and standard deviations for chest examinations were also determined. Dose values were compared with those used as references by the European Commission and the American Association of Physicists in Medicine (AAPM) (4,8).


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 ADVANCES IN KNOWLEDGE
 References
 
During the first 3 months of 1999 (beginning of operation for the CR system), an increase in patient doses was detected (Table 1). For the examination types monitored, significant dose reductions have been demonstrated over consecutive years since CR was implemented (Fig 1). With the dose values of 1999 used as the baseline, the Kruskal-Wallis test results showed significant global differences over the period of study (asymptotic P value < .001 in all cases). Applied to pairs of consecutive years, the Mann-Whitney test results showed that significant decreases in dosage occurred between 2001 and 2002 (P < .025 for all examinations, except thoracic spinal examinations) (Table 2).


View this table:
[in this window]
[in a new window]

  		Table 1. Mean Entrance Surface Doses before and after Installation of Digital Equipment

 

Figure 1
View larger version (9K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1: Graph of evolution of median values for abdominal, pelvic, and thoracic spinal examinations during period of study. After uneven results during beginning of operation of CR system in 1999, a slow decrease in median entrance dose values occurred for analyzed examinations from 1999 to 2001, with considerable decreases after training activities and routine QC operation of online audit system in 2001. AP = anteroposterior, LA = lateral.

 

View this table:
[in this window]
[in a new window]

  		Table 2. Yearly Median Entrance Surface Doses after the Introduction of CR

 
At present, our median entrance dose values range between 28% and 41% of the available reference values recommended by the AAPM (8) for conventional screen-film radiography, and between 15% and 38% of those recommended by the European guidelines document (4). The maximum differences in median values for patient doses were observed between 2001 and 2002. For example, median values for posteroanterior chest examinations decreased from 0.15 to 0.09 mGy in this period (Fig 2). For chest examinations, for which we aimed to compare our results with those of other authors who assume normal distributions, the mean doses (± standard deviations) were 0.20 mGy ± 0.09 in 1999 and 0.11 mGy ± 0.09 in 2003 for posteroanterior and 0.93 mGy ± 0.65 in 1999 and 0.64 mGy ± 0.50 in 2003 for lateral examinations. For a considerable number of examinations, kilovolt peak settings were concentrated at two main values, and SSDs showed important variations (eg, pelvic examinations in 2003) (Fig 3).


Figure 2
View larger version (18K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2: Histograms of dose distributions, including median (also plotted as a vertical line), first and third quartile values (in parentheses), mean, and standard deviation (Std. dev.) during 2001 (left) and 2002 (right) for chest imaging. Sample sizes indicate number of examinations. Furthest outlier values have not been shown.

 

Figure 3
View larger version (16K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 3: Kilovolt peak and focus-to-detector distance distributions for pelvic imaging in 2003, including means and standard deviations (Std. dev.); sample size indicates number of examinations.

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
ADVANCES IN KNOWLEDGE
 References
 
It is not surprising that average doses increased for some examinations when CR was introduced, as discussed and reported elsewhere (3); this increase was very apparent for chest and lumbar spinal examinations, the only examinations for which data were available before 1999. In our study, the staff had no or limited experience with digital technology, and the radiographers probably attempted to avoid noisy images by using milliampere-second settings higher than necessary for good image quality. Use of low kilovolt peaks to increase contrast and using short SSDs were typical mistakes made by radiographers, even after the training activities held in 2001 (a worse situation [more mistakes] than that in 2003 was found in earlier years). These points were discussed in a previous dose audit of screen-film radiography and detected during a survey of patient doses (9) performed at our institution as part of a European research program. The initial increase in median yearly dose values for pelvic and anteroposterior lumbar spinal imaging from 1999 to 2001 can be explained by the inadequate selection of technical parameters for exposure.

The effect of the kilovolt peak setting on the patient entrance dose at CR has been described by Lu et al (10), who suggested the use of higher kilovolt peak settings and additional filtration of 2 mm Al at CR to reduce the patient entrance dose without compromising contrast-detail detectability. Any decrease in contrast can be offset by manipulation of the window width and level settings on soft-copy display workstations. Gray (11) recommends using a half-value layer of 3.0 mm Al at 80 kVp for diagnostic imaging. Moreover, the recently published regulations from the U.S. Food and Drug Administration (12) have increased the minimum half-value layer at 80 kVp from 2.3 to 2.9 mm Al.

Saiani et al (13) obtained doses of 0.30 mGy ± 0.05 and 0.90 mGy ± 0.17 for posteroanterior and lateral chest examinations, respectively, as typical entrance surface doses at chest radiography performed with a CR system (FCR 5000R CLS system; Fujifilm Medical Systems, Tokyo, Japan). These values are higher than our means and standard deviations in both 1999 and 2003.

Throughout the dose reduction program, median values decreased by 50% and 37% between 1999 and 2003 for posteroanterior and lateral chest radiography, respectively. Dose reductions could be interpreted as a benefit of training conducted by the hospital's medical physics service and application of the online patient dose audit system. From 2003 to 2004, the trend for chest and lumbar spinal examinations changed, with a slight increase in doses, suggesting a need for more training activities, especially for radiographers who joined the department after 2001. At present, the median entrance dose values at the CR examinations monitored in our department are in the range of 15%–38% of the recommended European diagnostic reference levels for conventional screen-film radiography (4) and 28%–41% of the reference levels recommended by the AAPM in the United States (8). The online patient dosimetry system has proved useful in the dose reduction program.

Since the study involved examinations performed with undercouch or onwall stand Bucky units, the results apply only to these types of examinations. Other examinations were not monitored, and this was therefore one limitation of the study. The use of a standard patient thickness for entrance dose estimation limits the overall precision of the calculated entrance surface dose values, as it does not reflect the true skin dose for larger or smaller patients, as discussed elsewhere (5). The inaccuracy in the entrance dose should not exceed 15%, however, provided that the actual milliampere-second used for each image is sent to the computer to calculate entrance surface dose, assuming a reasonable range in patient thicknesses of ±5 cm around the mean for adults (children are not examined in these rooms), and given the focus-to-photostimulable-phosphor distances used at the examinations (usually 110 cm, but 180 cm for chest examinations). Such variations are unimportant, considering the sample sizes used and the fact that small and large thicknesses cancel each other out. A final weakness of the study was the fact that patients may have been imaged more than once over the course of several years, and their data would be correlated. However, this limitation is of relatively minor importance, as our sample sizes were large and our analysis concerned radiation dose per exposure and not per patient.

Although patient dose values for projection radiography can increase during the transition from conventional screen-film radiography to CR, dose management programs for digital techniques, specific training of radiographers, and frequent patient dose audits can improve practice while maintaining or reducing patient doses. Digital techniques allow diagnostically adequate images to be obtained with substantially lower patient doses than used for screen-film radiography. Our study results have demonstrated the efficacy of the ICRP recommendations, namely, appropriate training and frequent patient dose audits.


   ADVANCES IN KNOWLEDGE
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 ADVANCES IN KNOWLEDGE
References
 


   ACKNOWLEDGMENTS
 
The authors thank P. Zuluaga, MSc, PhD, for her help in the statistical analysis.


   FOOTNOTES
 

Abbreviations: AAPM = American Association of Physicists in Medicine • CR = computed radiography • ICRP = International Commission on Radiological Protection • QC = quality control • SSD = source-to-skin distance

Authors stated no financial relationship to disclose.

Author contributions: Guarantor of integrity of entire study, E.V.; study concepts/study design or data acquisition or data analysis/interpretation, all authors; manuscript drafting or manuscript revision for important intellectual content, all authors; manuscript final version approval, all authors; literature research, E.V., J.M.F.; statistical analysis, E.V., H.d.L.H.; and manuscript editing, all authors


   References
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 ADVANCES IN KNOWLEDGE
 References
 

  1. International Commission on Radiological Protection. Managing patient dose in digital radiology: a report of the International Commission on Radiological Protection. Ann ICRP 2004;34:1–73.[Medline]
  2. Peters SE, Brennan PC. Digital radiography: are the manufacturers' settings too high?—optimisation of the Kodak digital radiography system with aid of the computed radiography dose index. Eur Radiol 2002;12:2381–2387.[Medline]
  3. Weatherburn GC, Bryan S, Davies JG. Comparison of doses for bedside examinations of the chest with conventional screen-film and computed radiography: results of a randomized controlled trial. Radiology 2000;217:707–712.[Abstract/Free Full Text]
  4. European Commission. European guidelines on quality criteria for diagnostic radiographic images. Report EUR 16260. Office for the Official Publications of the European Communities. Published 1996. ftp://ftp.cordis.lu/pub/fp5-euratom/docs/eur16260.pdf. Accessed October 20, 2005.
  5. Vano E, Fernandez JM, Ten JI, Guibelalde E, Gonzalez L, Pedrosa CS. Real-time measurement and audit of radiation dose to patients undergoing computed radiography. Radiology 2002;225:283–288.[Abstract/Free Full Text]
  6. Applegate KE, Tello R, Ying J. Hypothesis testing III: counts and medians. Radiology 2003;228:603–608.[Abstract/Free Full Text]
  7. SPSS Web site. http://www.spss.com. Accessed October 20, 2005.
  8. Gray JE, Archer BR, Butler PF, et al. Reference values for diagnostic radiology: application and impact. Radiology 2005;235:354–358.[Abstract/Free Full Text]
  9. Vano E, Oliete S, Gonzalez L, Guibelalde E, Velasco A, Fernandez JM. Image quality and dose in lumbar spine examinations: results of a 5 year quality control program following the European quality criteria trial. Br J Radiol 1995;68:1332–1335.[Abstract/Free Full Text]
  10. Lu ZF, Nickoloff EL, So JC, Dutta AK. Comparison of computed radiography and screen-film combination using a contrast-detail phantom. J Appl Clin Med Phys 2003;4:91–98.[CrossRef][Medline]
  11. Gray JE. Quality control in diagnostic imaging: a quality control cookbook. Baltimore, Md: University Park Press, 1983.
  12. Performance standard for diagnostic x-ray systems and their major components, 21 CFR §1020.30, §1020.31, §1020.32, and §1020.33 (2005).
  13. Saiani F, Ghirardi C, Rodella CA, Feroldi P, Chiesa A. Radiation dose in digital chest radiography: comparison among three technologies. Radiol Med (Torino) 2004;107:401–407.[Medline]



This article has been cited by other articles:


Home page
 RadiologyHome page
F. A. Mettler Jr, W. Huda, T. T. Yoshizumi, and M. Mahesh
Effective Doses in Radiology and Diagnostic Nuclear Medicine: A Catalog
Radiology, July 1, 2008; 248(1): 254 - 263.
[Abstract] [Full Text] [PDF]

Home page
 Radiat Prot DosimetryHome page
J. M. Fernandez, J. M. Ordiales, E. Guibelalde, C. Prieto, and E. Vano
Physical image quality comparison of four types of digital detector for chest radiology
Radiat Prot Dosimetry, March 1, 2008; 129(1-3): 140 - 143.
[Abstract] [Full Text] [PDF]

Home page
 Radiat Prot DosimetryHome page
K. Smans, E. Vano, R. Sanchez, F. W. Schultz, J. Zoetelief, T. Kiljunen, C. Maccia, H. Jarvinen, R. Bly, A. Kosunen, et al.
Results of a European survey on patient doses in paediatric radiology
Radiat Prot Dosimetry, March 1, 2008; 129(1-3): 204 - 210.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
2432050930v1
243/2/461    most recent
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
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 Vaño, E.
Right arrow Articles by de Las Heras, H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Vaño, E.
Right arrow Articles by de Las Heras, H.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
RADIOLOGY RADIOGRAPHICS RSNA JOURNALS ONLINE