Thesis submitted in fulfilment of the requirements for the award of the degree of Master of Technology Radiography (Diagnostic) in the Faculty of Health and Wellness Sciences at the Cape Peninsula University of Technology 2009 / Since their discovery in 1895, the use of x-rays is continuously evolving in medicine making the diagnosis of injuries and diseases more practicable. It is therefore not surprising that x-rays contribute 90% of the radiation dose to the population from manmade sources (DWP, 1992). Moreover, these radiation doses are associated with both fatal and non-fatal cancer risk that is detrimental to adults between 20 to 60 years (Wall, 1996). Radiation dose to individuals therefore needs to be actively monitored in order to minimise such risk. Barium contrast examinations were characterised as one of the radiological examinations that contributed enormously to the collective dose to the patients in the radiology department (DWP, 1992). Determining the diagnostic reference levels of such examinations would reduce the over-exposure of individuals to ionising radiation. Currently in South Africa (SA), there are no diagnostic dose reference levels for barium meal (BaM) and barium enema (BaE) examinations. This study therefore investigated the radiation doses delivered to patients referred for BaM and BaE, obtained potential regional reference doses for these examinations, compared the radiation doses obtained with those from similar dosimetry studies and investigated sources of dose variation among the study sites. A total of 25 BaM and 30 BaE patients in the age range 18 to 85 years, weighing 50 kg to 90 kg, at 3 hospitals in the Western Cape, SA were investigated. The radiation dose to the patients was measured using Dose Area Product (DAP) meters that were permanently fitted onto fixed fluoroscopy units at these 3 hospitals. The third quartile DAP values were 20.1 Gycm2 and 36.5 Gycm2 for BaM and BaE respectively. The median DAP values were 13.6 Gycm2 and 27.8 Gycm2 for BaM and BaE respectively. The median values were recommended as the potential Diagnostic Reference Levels for BaM and BaE as they are less affected by outlying values of under or over- weight (Yakoumakis, Tsalafoutas, Sandilos, Koulentianos et al, 1999). The weights of the patients, fluoroscopy time, the number of images obtained, the use of digital or conventional fluoroscopy equipment and the level of training of the radiologists were the factors considered for dose variation among the 3 hospitals.

The International Commission on Radiological Protection (ICRP) determines the policy regarding radiation safety internationally. To the ICRP, hereditary changes as a result of either high or low doses, are of a major concern. The SA Forum for Radiation Protection recommended that a research project to determine the genetically-significant dose (GSD) for the South African population should be done as such a project has never been undertaken to date. This term was at first defined by UNSCEAR in 1958. The National Radiological Protection Board derived a formula from this definition as shown in the NRPB Report, NRPB-R106 (1980). This formula was implemented in the project. It combines the frequency of radiological examinations obtained during the country-wide survey and estimates of gonadal doses for different examination types, together with population and child expectancy data. New procedures, techniques and data processing that were relevant to this project had to be developed because the available information and conditions are unique to South Africa. The task was set to find a model in order to draw the best representative sample of the population and it was determined in a unique way, namely the so-called Dollar Unit Sampling method. A sample of 27 institutions out of a possible 292 (9%) was drawn in comparison, e.g., with the 8% of France and 8% in Great Britain. It was necessary to rely mainly on the calculation of gonad doses due to a shortage of manpower, contrary to other countries that were able to physically measure doses. Information obtained in the survey was used in this regard. The "RADCOMP Entrance Skin Exposure Software Program " of Nuclear Associates was used to produce parametric Free Air Exposure tables based on doses from Table B.3, NCRP Report No. 102. After the skin entrance doses were calculated, it was possible to calculate the gonad doses. A computer program was obtained from the Food and Drug Administration in the USA for this purpose. Data analysis was performed by means of the software package Microsoft Excel version 4.0. The above-mentioned formula was used in order to obtain the final results. The GSD for the total SA-population was calculated as 94.6 μGy. The breakdown of the GSD for the various South African race groups was Asian - 229.0 μGy, Black - 66.5 μGy, Coloured - 112.2 μGy and White - 463.4 μGy.