The Study: European guidelines on quality criteria for diagnostic radiographic images of the lumbar spine - an intra- and inter-observer reproducibility study

The Facts:

a. “The Commission of the European Communities has published guidelines to be used as a gold standard for quality assessment of diagnostic radiographic images. Image quality and radiation dose must be monitored and optimally balanced for diagnostic purposes on patients.”

b. There are 249 chiropractic clinics in Denmark. The vast majority of these clinics have their own x-ray facilities.

c. The Nordic Institute for Chiropractic and Clinical Biomechanics at the University of Southern Denmark set up a Picture Archiving and Communication System in 2008.

d. At this time the system has been expanded to the point where it has functions to document quality assurance and has a second opinion service.

e. It allows a monitoring of radiation exposure and image quality.

f. There is a quality assessment protocol that has been proposed by the Commission of the European Communities [CEC].

g. “The image criteria specify important anatomical structures that should be visible in a radiograph to aide accurate diagnosis.”

h. For this study they selected two doctors of chiropractic [the observers who rated the radiographs] who were in their first two years of clinical practice.

i. They were selected because the authors wished to study how use of the guidelines for quality assessment would work for less experienced clinicians.

j. “The observers received adequate introduction in the use of the CEC quality criteria”.

k. They divided the radiographs into what they judged to be accepted and nonaccepted radiographs as regards to x-ray quality.

l. The clinicians looked at 50 lumbar radiographs and then 4 weeks later they looked at them again and determined which ones were acceptable and which were nonacceptable.

m. “According to the methods recommended by Landis and Koch, our results can be rated with moderate to substantial agreement.”

n. “The inter-observer and intra-observer agreements in this study, using the CEC-criteria, were found to be mostly acceptable.”

o. The authors felt more experienced observers might be able to do a better job.

p. “Our results indicate that less experienced observers likely would benefit from training in an initial trial of at least 50 imaging studies.”

Take Home:

You can train clinicians to make decisions on whether a lumber radiograph has an acceptable level of quality [such as being able to see certain structures].

Reviewer's Comments:

I think that we should all be concerned about the quality of radiographs. The better the quality, the more accurate the information we have to use in caring for the patient. You will note that there is not complete agreement between the two clinicians who did the evaluations. That is to be expected as this type of evaluation is an acquired skill even if you have criteria that you are using. Of course in a real practice setting we would hope that there would be allowances made for real world clinical events such as it is easier to get a good radiograph of a six foot tall 160 pound patient as opposed to a five foot tall 320 pound patient and we don’t want to inappropriately exclude some patients from receiving necessary tests just because the images may not make the doctor look as good to the observer [rater]. These types of studies are a good idea. I welcome improvements in all sorts of clinical parameters and applaud the authors even though I expect that someone will use this study inappropriately. Does that sound a little too cynical? Forgive me. I’m old. I’ve just seen too much.

Reviewer: Roger Coleman DC

 Editor: Mark R. Payne DC

 Reference: Doktor K, Vilholm ML, Hardardottir A, Christensen HW, Lauritsen J, Doktor K. European guidelines on quality criteria for diagnostic radiographic images of the lumbar spine - an intra- and inter-observer reproducibility study. Chiropr Man Therap. 2019 May 1;27:20. doi: 10.1186/s12998-019-0241-3. eCollection 2019.

 Link to Abstract: https://www.ncbi.nlm.nih.gov/pubmed/31069046