Chapter 2

The concept of quality criteria for diagnostic x-ray examinations was introduced by the CEC as European Guidelines on 'Quality Criteria for Diagnostic Radiographic Images' (1).These provide guidance on diagnostic requirements, the radiation dose to the patient and the choice of radiographic technique. The concept was subsequently applied to paediatric radiology (2) and, in this report, is now being extended to CT. In the meantime evaluations of the application of quality criteria have been performed for conventional radiography images and CT, as summarised below.

Quality criteria for conventional diagnostic radiographic images in adult radiology have been developed over a period of about ten years during which two European-wide trials have been conducted in order to assess relevance, acceptability and ease of use for technical and clinical staff in diagnostic x-ray departments. The findings of the trials provide a supplementary scientific background to the application of such quality criteria (3,4).

The first European trial (3) was conducted in 1987/88 and involved information from 24 x-ray departments in 10 European countries. It concerned radiographic technique and compliance with the image criteria given in the preliminary Quality Criteria Working Document. The results confirmed the validity of the application of quality criteria as a tool for the optimisation of radiation protection. In particular they permitted the identification of suitable technical modalities for achieving the best possible compromise between the essential medical information in a radiographic image and the patient dose. However, the trial clearly highlighted the need for establishing quality assurance programmes and quality control protocols in diagnostic radiology, since large variations in dose were found for the same type of x-ray examination.

In order to assess the validity of a revised Quality Criteria Working Document and to overcome some of the limitations of the first trial, a second trial was carried out in 1991(4). Original films of chest, lumbar spine and breast radiographs were sent to an independent panel of radiologists for assessment against the image criteria. A questionnaire was employed to collect entrance surface doses and details of the radiographic equipment and technique factors in use. The results of the 1991 trial highlighted a number of important features of radiographic practice in Europe. Information concerning the technical parameters of radiographic equipment was still in many cases not sufficiently known by staff in x-ray departments. The trial demonstrated that in conventional radiology the entrance surface dose provides a useful measure of the patient dose and it confirmed that there existed wide variations in performance throughout Europe. Furthermore the results showed that radiologists find it difficult to interpret unequivocally criteria which involve some form of assessment of symmetry, field coverage and fulfilment of technical requirements. It was concluded that a compartmentalisation of image criteria was required in order to link, more consistently, the image quality, patient dose and radiographic technique.

Besides such European wide trials, reports are also becoming available concerning the experiences of clinical application of the quality criteria for conventional radiography at a national level. For example, Vaño and colleagues (5,6) have confirmed the clinical applicability of the quality criteria concept and illustrated the potential for dose reduction in Spain.

To date no results of research studies are available to indicate to what extent radiation dose may be reduced while retaining clinical effectiveness of the examination. The basis for the current guidelines is the supposition that the quality criteria concept, which was developed for conventional radiography, can also be used for advancing optimisation of CT examinations. Nevertheless some fundamental modifications have had to be made owing to the particular characteristics of CT, for example, patient dose during CT should not be expressed as entrance surface dose but in terms of quantities that have been specifically developed for CT, such as the weighted computed tomography dose index (CTDI_W) for a slice and the dose-length product (DLP) for a complete examination.

For the establishment of the quality criteria for CT extensive data were currently available on patient dosimetry and patient dose in relation to CT examinations (7-13). In addition, some information has been published concerning the relationship between dose and CT image quality (14-16). More comprehensive approaches to patient dose, diagnostic image quality and technical parameters have also been described for CT (17-19).

The elaboration of the image criteria in CT has been difficult due to the complexity of CT anatomy and technique. The image criteria that have been elaborated by the German Federal Chamber of Physicians (20) proved to be a valuable source of information. They were tested in a pilot study at the University Hospital of Aarhus for CT of the mediastinum (23 examinations) and CT of the retroperitoneal space (30 examinations) (17). The analysis showed that it is necessary to differentiate the degree of visualization of the anatomical structures by including the term visually sharp reproduction instead of just visualization of critical anatomical structures. Consequently a set of modified criteria was tested in the same institution (190 examinations) (17) and a new list of criteria was elaborated for the mediastinum and the retroperitoneal space. The results showed that nearly all of the new criteria were useful for measuring the diagnostic image quality as they were fulfilled in an acceptable amount of examinations but not always (18). These criteria were included in the first working document on CT quality criteria that was published April 1997. It was sent out for comments to professional groups, such as radiologists, radiographers, and medical physicists as well as to manufacturers and health care authorities in the member states of the European Union. This request yielded some 50 responses. The quality criteria and reference doses for CT were presented at several congresses and symposia, e.g. at ECR'97 and at the EC Workshop on Reference Doses and Quality in Medical Imaging, held October 1997 in Luxembourg. At this Workshop the University of Aarhus presented results and guidelines on image criteria for CT brain, based on 119 examinations (21). Also the first experiences with the clinical implementation of quality criteria for 102 CT brain examinations were presented by Calzado et al. and found to be useful (22).

The comments on the first working document were incorporated in the May 1998 version which was discussed at the EC-workshop on quality criteria for computed tomography, held November 1998 in Aarhus, Denmark. The workshop aimed at deriving consensus on the 1998 working document before publication. Of the 49 participants, 46 came from 12 European countries, 2 from Brazil and 1 from the USA. They represented hospitals, governments, professional bodies and manufacturers of CT scanners.

At the workshop and at ECR'99 results of a pilot trail were presented, which evaluated the quality criteria for examination of the face and sinuses, the spine, the chest (HRCT), liver and spleen, and osseous pelvis in hospitals of four countries (Denmark, The Netherlands, Switzerland and the United Kingdom).This trail showed that high radiation doses to the patient did not always imply optimal diagnostic quality. Dose reduction, especially regarding examinations of the face and sinuses and osseous pelvis, seemed to be achievable without loss of diagnostic image quality. It was concluded that the quality criteria can be used to optimise these CT procedures. As a result of this study some changes of the criteria for the liver and spleen and for osseous pelvis were included in the present guidelines.