British Journal of Radiology 74 (2001),255-258 © 2001 The British Institute of Radiology
Influence of radiologists' sex and training on fluoroscopy doses during barium enema
T Vehmas, MD
1,2 and
E Kuosma, MSc
1
1 The Finnish Institute of Occupational Health, Topeliuksenkatu 41 a A, Helsinki, FIN-00250 Helsinki
2 The Department of Radiology, Helsinki University Central Hospital, Finland
Correspondence: Dr T Vehmas, The Finnish Institute of Occupational Health, Topeliuksenkatu 41 a A, FIN-00250 Helsinki, Finland
 |
Abstract
|
|---|
Variables influencing radiation exposure were studied using analysis of covariance in 523 double contrast barium enemas performed by 21 radiology residents (12 female, 9 male). Dose–area products (DAPs) (ns), number of exposures per patient (p<0.05) and screening times (p<0.05) were higher for female residents (mean 52.0 Gy cm2, 9.55 exposures and 4.02 min, respectively) than for male residents (mean 41.7 Gy cm2, 8.26 exposures and 3.20 min, respectively). The number of lectures on radiation protection attended by the resident (range 0–20) had no significant effect on these radiation-related variables. Other factors explaining these variables were patient age, diagnosis and anteroposterior thickness. Radiologists' use of ionizing radiation during fluoroscopy may be sex-related. Traditional lecture-based education on radiation protection was not effective with this group; thus, improved methods should be developed to control doses.
|
Introduction
|
|---|
Barium enema is currently the most common fluoroscopic procedure in Canada [1], although its frequency has been diminishing from 1985–1992, when 511 barium enemas per 100 000 persons were performed in the United States [2]. Adult digestive tract fluoroscopy contributed 5–17% of the annual collective dose in a large Spanish hospital during 1986–1990 [3]. In Finland, barium enemas account for 13.9% (330 manSv) of the annual collective effective dose [4]. Previous reports in several countries have shown up to 10-fold variation in the mean effective doses from barium enemas [5–9]. If protection was optimized, then it is estimated that 7500 manSv could be saved in the UK, potentially saving between 100 and 250 out of 160 000 deaths from cancer annually [10]. The radiation dose in a fluoroscopic study is dependent on variables related to the patient, the equipment and the examiner. Patients cannot usually be selected, nor can the equipment after installation. The examiner, on the other hand, can control the use of fluoroscopy as well as the number of exposures during the study. The aim of this work was to study the influence of the examiner on radiation exposure during barium enema. Radiation-related variables (dose–area product (DAP), screening time and number of exposures per patient) were measured. Examiners were radiology residents; their sex, radiological experience and number of radiation protection lectures attended previously were correlated with these radiation-related variables.
|
Materials and methods
|
|---|
A total of 523 double contrast barium enemas were studied between October 1996 and July 1998. There were 12 female residents and 9 male residents performing these examinations. Each resident conducted between 9 and 33 studies (mean 25). The residents completed a questionnaire regarding their age, the duration of their radiological career, the estimated number of barium enemas performed previously (<100, 100–500, >500) and the number of radiation protection lectures attended.
Patients participating in this study (332 female, 191 male, age range 19–97 years, mean 62 years) were unselected and had various indications forbarium enema. Before imaging, bowel preparation was carried out. A standard double contrast technique was used. Radiographs were exposed using Shimadzu fluoroscopic equipment (Shimadzu, Kyoto, Japan) with an undercouch tube and two translateral radiographs using separate equipment (the latter dose did not contribute to DAP). After processing the digital plates (Fuji ST V) with an FCR AC-2 unit, hard copies were printed with an FCR 700 laser image printer (Fujifilm, Tokyo, Japan). The radiographs were then inspected (usually under the supervision of a senior radiologist, who was responsible for the diagnostic process) and, in unclear cases, the study was continued with further fluoroscopy and radiographs. Screening times in the Shimadzu control panel were recorded. DAPs were measured with a meter (DAP-S, Model 841-S, Gammex-RMI Inc., Middleton, USA).
Another questionnaire was completed after every examination. This included the patient's sex, age, weight and height; the particular screening time and DAP; the number of films exposed with the screening device; and the main radiological diagnosis (normal, diverticulosis, polyp(s) or suspicion, tumour or suspicion, other). The group "other" included post-operative states, a combination of diagnoses judged equally important and miscellaneous diagnoses.
Statistical methods
The SAS program [11] was used to calculate a linear mixed analysis of covariance (ANCOVA) model separately on DAP, screening time and number of exposures per patient. Resident was used as a random coefficient in the models. Since the distributions of DAP and screening time were skewed, a natural logarithmic transformation was applied. The variables in the models were patients' sex, age, anteroposterior (AP) thickness (4.22+39.81xweight [kg]/height [cm]; only used for DAP) [12] and diagnosis, and residents' sex, number of previously performed barium enemas (classified: <100, 100–500, >500), duration of their radiology career and number of lectures on radiation protection attended. First, all reasonable independent variables were included in the models. Intercorrelated variables and variables with poor p-values were then excluded. Several alternative models were computed.
|
Results
|
|---|
At the beginning of the study, after six residents had performed 75 barium enemas, the screening times were compared to pre-study screening times that had been continuously and unnoticeably recorded by the radiographers. Screening times prior to the study (mean 5.2 min) were significantly longer than during the first 75 study examinations (mean 3.6 min) [13]. The screening times remained at a reasonably low level (mean 3.66 min) throughout the study. DAP ranged from 8.83 Gy cm2 to 177 Gy cm2 (mean 47.5 Gy cm2).
A comparison between female and male residents is given in Table 1
. The mean DAP for female residents was higher (52.0 Gy cm2) than for male residents (41.7 Gy cm2). This was due both to their longer screening times (mean 4.03 min) compared with males (3.20 min) and the greater number of exposures per examination (9.55 vs 8.26). The mean AP thickness of patients examined by female radiologists was significantly greater (21.4 cm) than that of patients studied by male radiologists (20.7 cm; p=0.01, t-test). The pathological diagnoses did not differ significantly for male and female radiologists (
2=3.51, df=4, p=0.48, chi-square test).
A normal diagnosis (Table 2) generally predicted low radiation values while higher values were encountered in pathological states, especially in (suspected) tumour cases.
View this table:
[in this window]
[in a new window]
|
Table 2. Effect of diagnosis on radiation-related variables (mean values): dose–area product (DAP), screening time and number of exposures per patient
|
|
ANCOVA results
The patient's AP thickness was the only significant predictor of DAP (positive correlation), while the patient's diagnosis reached a borderline significance (Table 3). Screening time (Table 4) was significantly affected by the patient's age (positive correlation) and diagnosis as well as by the radiologist's sex (which becomes significant after removing the least important variables from the model). The number of exposures (Table 5) was significantly determined only by the radiologist's sex. The number of lectures on radiation protection attended had no significant effect on any radiation-related variables (Tables 3–5). There was an insignificant positive correlation between all radiation-related variables and the number of these lectures attended.
View this table:
[in this window]
[in a new window]
|
Table 5. Covariance analysis of the number of exposures per barium enema examination: explanatory variables with their contribution (p-values)
|
|
|
Discussion
|
|---|
The current literature [14] has paid attention to the influence of the radiologist's technique on patient dose in barium studies. The most striking finding in our study was the dependence of screening time and the number of exposures on the examiner's sex. Female residents used more radiation than male residents. Does this finding suggest that women are more conscientious, work more carefully, or are less sure of themselves and therefore have to use more radiation? The difference between radiation parameters for male and female radiologists was far less than the difference between screening times before and during the study. Also, differences in radiation variables within sexes were striking, suggesting that individual behavioural patterns may be crucial.
The radiologists and patients could not be randomized owing to practical reasons. Each radiologist was supposed to examine as many successful barium enema patients as possible (up to 25–30) during his/her stay at the gastrointestinal radiology unit. Male radiologists may have succeeded in avoiding some obese patients as the mean AP diameter of their patients was significantly lower. Did male residents try to keep away from the most difficult cases? Including the patients' AP diameter in ANCOVA models had no major effect on the screening time and the number of exposures analyses. Adding the interaction term "radiologist's sexxpatient's AP diameter" also had no major effect on the DAP results. The patients' AP diameter was used to describe patients' body constitution because it was found to correlate better with DAP than patients' mass [kg] or body mass index.
It is likely that these sex differences will also affect the examiner's behaviour in other fluoroscopic studies. Further studies should be undertaken to work out character traits that may affect this radiation behaviour. Whether the reduced use of radiation adversely influences diagnosis (e.g. causes false negative/positive diagnoses) would also be important to determine and this cannot be judged from the results of this study.
The influence of the radiologist's grade on fluoroscopic patient dose has been studied by Hoskins and Williams [12]. They found a trend for both screening time and patient dose to fall to a minimum for third year resident/first year senior resident grades. The parameters rose from this group to the consultant level. The current multivariate study with simultaneous control of many patient- and radiologist-related intervening variables and with limited variation between the junior radiologists' experience cannot confirm this earlier result.
Contrary to what could be expected, attending a greater number of lectures on radiation protection had no significant effect on the use of ionizing radiation. This finding should be confirmed with a greater number of radiologists if possible. Traditional lecture-based information on radiation protection was not an effective way to control doses in this group. More pragmatic fluoroscopy credentialling and practice surveillance has been effective in controlling screening times [15].
|
Conclusion
|
|---|
Unlike DAP, screening times and the number of exposures were increased for female radiologists, indicating sex-related fluoroscopy behaviour. Traditional lecture-based education on radiation protection is not an effective way to control doses and better methods should be developed.
|
Acknowledgments
|
|---|
We wish to thank the residents and radiographers for their help with data collection, and Ms Terttu Kaustia, MA, for the language revision.
|
Footnotes
|
|---|
The work was supported by the P. O. Klingendahl foundation 
Received for publication November 12, 1999.
Revision received September 11, 2000.
Accepted for publication October 16, 2000.
|
References
|
|---|
-
Lavoie C, Don C. In vivo measurement of ovarian dose during barium enema examinations. Br J Radiol 1997;70:291–5.[Abstract]
-
Karasick S, Ehrlich SM, Levin DC, Harford RJ, Rosetti EF, Ricci JA, et al. Trends in the use of barium enema examination, colonoscopy and sigmoideoscopy: is use commensurate with risk of disease? Radiology 1995;195:777–84.[Abstract/Free Full Text]
-
Vañó E, Velasco A, Morán P, González L, Pedrosa CSA. Evolution of diagnostic radiology in a big hospital during a 5 year period, and the derived collective dose. Br J Radiol 1993;66:892–8.[Abstract/Free Full Text]
-
Rannikko S, Karila KTK, Toivonen M. Patient and population doses of X-ray diagnostics in Finland, STUK-A144. Helsinki: Radiation and Nuclear Safety Authority, 1997.
-
Hart D, Jones DG, Wall BF. Estimation of effective dose in diagnostic radiology from entrance surface dose and dose–area product measurements, NRPB-R262. Chilton: NRPB, 1994.
-
Geleijns J, Broerse JJ, Shaw MP, Schultz FW, Teeuwisse W, van Unnik JG, et al. Patient dose due to colon examination: dose assessment and results from a survey in the Netherlands. Radiology 1997;204:553–9.[Abstract/Free Full Text]
-
Martin CJ, Hunter S. Reduction of patient doses from barium meal and barium enema examinations through changes in equipment factors. Br J Radiol 1994;67:1196–205.[Abstract/Free Full Text]
-
Bauer B, Veit R. Initiatives, achievements and perspectives in quality assurance and radiation protection in diagnostic radiology, both on legal and practical level in Germany. Radiat Protect Dosim 1995;57:43–6.
-
Staniszerwska MA. X-ray diagnostics as the main source of ionising radiation for the general Polish population. J Radiat Prot 1993;13:275–8.
-
NRPB. Report by the Royal College of Radiologists and the National Radiological Protection Board. Patient dose reduction in diagnostic radiology, Documents of the NRPB, Vol. 1, No. 3. London: NRPB, 1990.
-
Littell RC, Milliken GA, Stroup WW, Wolfinger RD. SAS system for mixed models. Cary, NC: SAS Institute Inc, 1996:253–66.
-
Hoskins PR, Williams JR. Influence of radiologist grade on fluoroscopic patient dose. Br J Radiol 1992;65:1119–23.[Abstract/Free Full Text]
-
Vehmas T. Hawthorne effect: shortening of fluoroscopy times during radiation measurement studies. Br J Radiol 1997;70:1053–5.[Abstract]
-
Corbett RH. The influence of radiologist's technique on patient dose in barium studies. In: Proceedings of the Tenth International Congress of the International Radiation Protection Association; 2000 May 14–19; Hiroshima, Japan. Paris, France: IRPA, 2000.
-
McParland BJ, Lewall DB. Reductions in fluoroscopy screening times resulting from physician credentialling and practise surveillance. Br J Radiol 1998;71:461[Medline]
This article has been cited by other articles:
|
|
|
|
 
W. M. McCartney
Radiation Dose During Barium Enema
Radiol. Technol.,
September 1, 2007;
79(1):
91 - 93.
[Full Text]
[PDF]
|
|
|
|
|
|
|
T Vehmas
Do radiologists' radiation-related opinions predict their fluoroscopy doses?
Br. J. Radiol.,
September 1, 2003;
76(909):
653 - 655.
[Abstract]
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
