To Örebro University

Objectives: Abdominal radiography is frequently used in acute abdominal non-traumatic pain despite the availability of more advanced diagnostic modalities. This study evaluates the diagnostic accuracy of low-dose CT compared with abdominal radiography, at similar radiation dose levels.

Methods: Fifty-eight patients were imaged with both methods and were reviewed independently by three radiologists. The reference standard was obtained from the diagnosis in medical records. Sensitivity and specificity were calculated. A systematic review was performed after a literature search, finding a total of six relevant studies including the present.

Results: Overall sensitivity with 95 % CI for CT was 75 % (66-83 %) and 46 % (37-56 %) for radiography. Specificity was 87 % (77-94 %) for both methods. In the systematic review the overall sensitivity for CT varied between 75 and 96 % with specificity from 83 to 95 % while the overall sensitivity for abdominal radiography varied between 30 and 77 % with specificity 75 to 88 %.

Conclusions: Based on the current study and available evidence, low-dose CT has higher diagnostic accuracy than abdominal radiography and it should, where logistically possible, replace abdominal radiography in the workup of adult patients with acute non-traumatic abdominal pain.

Key points: • Low-dose CT has a higher diagnostic accuracy than radiography. • A systematic review shows that CT has better diagnostic accuracy than radiography. • Radiography has no place in the workup of acute non-traumatic abdominal pain.

Background: Lumbar spine radiography has limited diagnostic value but low radiation dose compared with computed tomography (CT). The average effective radiation dose from lumbar spine radiography is about 1.1 mSv. Low-dose lumbar spine CT may be an alternative to increase the diagnostic value at low radiation dose, around 1 mSv.

Purpose: To determine the optimal settings for low-dose lumbar spine CT simultaneously aiming for the highest diagnostic image quality possible.

Material and Methods: An ovine lower thoracic and lumbar spine phantom, with all soft tissues around the vertebrae preserved except the skin, was placed in a 20 L plastic container filled with water. The phantom was scanned repeatedly with various technical settings; different tube potential, reference mAs, and with different convolution filters. Five radiologists evaluated the image quality according to a modification of the European guidelines for multislice computed tomography (MSCT) quality criteria for lumbar spine CT 2004. In a visual comparison the different scans were also ranked subjectively according to perceived image quality. Image noise and contrast were measured.

Results: A tube potential of 120 kV with reference mAs 30 and medium or medium smooth convolution filter gave the best image quality at a sub-millisievert dose level, i.e. with an effective dose comparable to that from lumbar spine radiography.

Conclusion: Low-dose lumbar spine CT thus opens a possibility to substitute lumbar spine radiography with CT without obvious increase in radiation dose.

Background: Lumbar spine radiography is often performed instead of CT for radiation dose concerns.

Purpose: To compare image quality and diagnostic information from low dose lumbar spine CT at an effective dose of about 1 mSv with lumbar spine radiography.

Material and Methods: Fifty-one patients were examined by both methods. Five reviewers scored all examinations on eight image quality criteria using a five-graded scale and also assessed three common pathologic changes.

Results: Low dose CT scored better than radiography on the following: sharp reproduction of disc profile and vertebral end-plates (odds ratio [OR], 1.8; 95% confidence interval [CI], 1.3-2.5), intervertebral foramina and pedicles (OR, 4.3; 95% CI, 3.1-5.9), intervertebral joints (OR, 139; 95% CI, 59-326), spinous and transverse processes (OR, 7.0; 95% CI, 4.3-11.2), sacro-iliac joints (OR, 4.2; 95% CI, 3.2-5.7), reproduction of the adjacent soft tissues (OR, 2.9; 95% CI, 2.1-4.0), and absence of any obscuring superimposed gastrointestinal gas and contents (OR, 188; 95% CI, 66-539). Radiography scored better on sharp reproduction of cortical and trabecular bone (OR, 0.3; 95% CI, 0.2-0.4). The reviewers visualized disk degeneration, spondylosis/diffuse idiopathic skeletal hyperostosis (DISH) and intervertebral joint osteoarthritis more clearly and were more certain with low dose CT. Mean time to review low dose CT was 204 s (95% CI, 194-214 s.), radiography 152 s (95% CI, 146-158 s.). The effective dose for low dose CT was 1.0-1.1 mSv, for radiography 0.7 mSv.

Conclusion: Low dose lumbar spine CT at about 1 mSv has superior image quality to lumbar spine radiography with more anatomical and diagnostic information.