Skip to main content
Top

Diagnostic accuracy of ultra-low-dose chest CT vs chest X-ray for acute non-traumatic pulmonary diseases

Published in:

Abstract

Objectives

To compare the diagnostic accuracy of ULDCT to CXR for detecting non-traumatic pulmonary diseases at the emergency department (ED) and to study diagnostic confidence levels.

Methods

Secondary analysis of the prospective OPTIMACT trial (2418 ED participants randomly allocated to ULDCT or CXR). Diagnoses at imaging at the ED were compared to the reference diagnosis on day 28. Ratios of positive diagnoses, true positives (TP), false positives (FP), false negatives (FN), and positive predictive values (PPV) were assessed with 95% confidence intervals (CI). The diagnostic confidence levels of the radiologists were studied.

Results

One thousand one hundred sixty-one ULDCT participants (mean age, 59 years ± 18 [standard deviation], 587 female) and 1151 CXR participants (mean age, 59 years ± 18 [standard deviation], 561 female) were evaluated. With ULDCT, pneumonia was 1.55 times (95% CI: 1.33–1.80) more often diagnosed at imaging at the ED, with significantly more TP (ratio 1.50; 95% CI: 1.26–1.76) and fewer FN (0.61; 95% CI: 0.37–0.99) but more FP (1.75; 95% CI: 1.19–2.58); a similar pattern was observed for other lower respiratory tract infections (LRTI). Pulmonary congestion was less often observed with ULDCT (0.45; 95% CI: 0.34–0.61), with fewer TP (0.50; 95% CI: 0.34–0.73), and FP (0.40; 95% CI: 0.24–0.65). PPVs were not significantly different. With ULDCT, radiologists were more often certain in diagnosing pneumonia (ULDCT 121/324, 37% vs CXR 48/208, 23%), LRTI (84/192, 44% vs 18/63, 29%), and no established disease (350/382, 92% vs 447/544, 82%).

Conclusion

Compared to CXR, ULDCT led to more TP but also more FP in detecting pneumonia and LRTI, while fewer TP and FP were found for pulmonary congestion. PPVs were comparable.

Key Points

Question Is ultra-low dose CT (ULDCT) more accurate than chest X-ray (CXR) for identifying non-traumatic pulmonary diseases in patients presenting at the ED?
Findings ULDCT detects more pulmonary infections in patients presenting at the ED with non-traumatic pulmonary complaints, while CXR detects more pulmonary congestion.
Clinical relevance ULDCT is superior to CXR in detecting pneumonia and other LRTI in ED patients, while CXR is superior in detecting pulmonary congestion. ULDCT can be an alternative for CXR in a selected group of patients.

Graphical Abstract

Title
Diagnostic accuracy of ultra-low-dose chest CT vs chest X-ray for acute non-traumatic pulmonary diseases
Authors
Maadrika M. N. P. Kanglie
Inge A. H. van den Berk
Tjitske S. R. van Engelen
Shandra Bipat
Patrick M. M. Bossuyt
Jan M. Prins
Jaap Stoker
on behalf of the OPTIMACT study group
Publication date
29-01-2025
Publisher
Springer Berlin Heidelberg
Keyword
Pneumonia
Published in
European Radiology / Issue 8/2025
Print ISSN: 0938-7994
Electronic ISSN: 1432-1084
DOI
https://doi.org/10.1007/s00330-024-11223-3
This content is only visible if you are logged in and have the appropriate permissions.
This content is only visible if you are logged in and have the appropriate permissions.

Next-generation MRI contrast agents: preparing the field (Link opens in a new window)

New MRI contrast agents are reshaping diagnostic imaging, promising lower gadolinium exposure amid evolving practice guidelines. How can you optimise contrast selection, dosing, and patient care in this rapidly advancing field?

This content is intended for healthcare professionals outside of the UK.

Independent Medical Education Grant:
  • Bayer HealthCare Pharmaceuticals Inc.
Learn more Link opens in a new window

How can your team use biomarkers to improve management of AD? (Link opens in a new window)

1.75 AMA PRA Category 1 Credits

Explore using biomarker testing to interpret results and guide treatment strategies through an interactive patient simulation and unique educational escape rooms.

This content is intended for healthcare professionals outside of the UK.

Independent Medical Education Grant:
  • Lilly
Learn more Link opens in a new window
Image Credits
Abstract graphic of layered, concentric circular shapes in bright green, pink, blue, and purple on a dark blue background. The rings and segments form a complex radial pattern without text/© Springer Health+ IME, Alzheimer’s disease in the biomarker era: preparing the multidisciplinary care team/© Springer Healthcare IME