Ultra-Low Dose CT Shows Promise as First-Line Imaging for Trauma

A multicenter study across five French hospitals shows that ultra-low dose (ULD) CT scans detect fractures more accurately that standard X-rays, potentially redefining first-lime imaging in emergency care. 

“The diagnostic performance of ultra-low dose CT is very close to standard CT, which was surprising to us,” noted Oumaya Ben Said, MD, who presented the research Wednesday morning. “This approach improves patient care while keeping radiation exposure near that of a conventional X-ray.”

The study was led by Joël Greffier, PhD, a medical physicist, and Maxime Pastor, MD, a radiologist, both from the Department of Medical Imaging at Nimes University Hospital in France. Their team began implementing ULD-CT acquisitions into routine clinical practice in 2015, initially focusing on chest imaging for patients with acute dyspnea in the emergency department (ED). 

Building on their early success, the team extended the acquisitions to osteoarticular imaging, aiming to investigate fractures in patients presenting to the ED with trauma, setting the stage for a comprehensive evaluation.

Collaboration Underscores Multidisciplinary Approach

With this foundation in place, the study prospectively enrolled 533 adults presenting to EDs with suspected fractures of the thoraco-lumbar spine, pelvis, proximal femur or extremities between October 2019 and September 2023. Each patient underwent three imaging exams: standard-dose CT (STD-CT), ULD-CT and standard radiography (STD-RX). Two radiologists independently recorded fracture locations and numbers, each blinded to the other exams.

The study yielded compelling results regarding the accuracy and reliability of ULD-CT compared to traditional methods. According to Dr. Ben Said, 282 patients (53%) had at least one fracture. “Compared with standard-dose CT considered as the reference, ultra-lowdose CT achieved a sensitivity of 95% and specific of 98%, missing only 13 fractures and producing six false positives,” Dr. Ben Said reported. 

In contrast, standard radiography had a sensitivity of 71% and specificity of 96%, with 81 false negatives and nine false positives.

The findings were further underscored by radiologists’ assessments of image quality and diagnostic confidence. “97% of exams were fully or mostly interpretable, with 91% considered high to excellent in diagnostic confidence,” Dr. Ben Said noted. “For standard radiography, interpretability was 83% and diagnostic confidence only 50%.”

ULD-CT’s accuracy was observed across fracture types, including spine, pelvis and extremities, with only minor limitations in hand, wrist and foot imaging. “We expected ULD-CT to outperform X-ray, but its sensitivity and image quality approaching that of standard-dose CT exceeded our expectations,” Dr Ben Said explained.

The study also measured interpretation time, which remained reasonable for ULD-CT despite the higher image volume. “Interpretation time is longer than standard X-ray (1.3 min vs 0.7 min), but ULD-CT can prevent repeat imaging when X-rays are inconclusive, saving time and resources in patient care,” Dr. Ben Said added.

This collaboration underscores the multidisciplinary approach necessary for implementing advanced imaging protocols in clinical practice. Recognizing the potential for broader application, the researchers plan to conduct a medical-economic study assessing the value of ULD-CT as a replacement for standard radiography in emergency trauma care. They also aim to explore extending ULD protocols to spectral and photon-counting imaging technologies.

“This research confirms that ultra-low-dose CT is a compelling alternative to standard radiography for trauma patients,” Dr. Ben Said emphasized. “It offers high diagnostic confidence and near-gold-standard performance, while maintaining low radiation exposure, which can enhance patient care and optimize department workflow.”

Access the session, “Multicentric Prospective Study Comparing Ultra-Low-Dose Ct, Standard Radiography and Standard-Dose Ct for Detecting Lesions in Patients Following Thoraco-Lumbar Spine, Pelvic, or Extremity Trauma,” (W3-SSER02-4) on demand at RSNA.org/MeetingCentral.