Imaging quality and diagnostic reliability of low-dose computed tomography lumbar spine for evaluating patients with spinal disorders

Spine J. 2014 Nov 1;14(11):2682-90. doi: 10.1016/j.spinee.2014.03.007. Epub 2014 Mar 17.


Background context: Computed tomography (CT) scans of the lumbar spine (CTLS) have demonstrated a higher level of accuracy than plain films and have been used to assess patients with spinal disorder when magnetic resonance imaging is not available. Nevertheless, radiation exposure remains a serious safety concern. Iterative reconstruction (IR) decreases the CT radiation dose for diagnostic imaging. However, the feasibility of using IR in CTLS is unclear.

Purpose: To evaluate the imaging quality and diagnostic reliability of CTLS with IR.

Study design: A prospective study.

Patient sample: All patients from outpatient departments who suffered from spinal disorders and were referred for CTLS.

Outcome measures: In acquired CT images, the signal-to-noise ratio (SNR) of the dural sac (DS), intervertebral disc (IVD), psoas muscle (PM), and L5 vertebral body, the contrast-to-noise ratio between the DS and IVD (D-D CNR), and the subjective imaging qualities were compared across groups. Interobserver agreement was evaluated with kappa values.

Methods: Patients receiving low radiation CTLS were divided into three groups. A 150 mAs tube current with 120 kVp tube voltage was used with Group A and a 230 mAs tube current with 100 kVp tube voltage with Group B. Intended end radiation exposure was 50% less than that of the control group. Tube modulation was active for all groups. The images of the two low-radiation groups were reconstructed by IR; those of the control group by filtered back-projection (FBP).

Results: The SNRs of the DS, IVD, PM, BM, and D-D CNR of Group A were not inferior to those of the control group. All SNRs and D-D CNRs for Group B were inferior to those of the control group. Except for that of the facet joint, all subjective imaging ratings for anatomic regions were equivalent between Groups A and B. Interobserver agreement was highest for the control group (0.72-0.88), followed by Group A (0.69-0.83) and B (0.55-0.83).

Conclusions: Fifty percent tube current reduction combined with IR provides equivalent diagnostic accuracy and improved patient safety when compared with conventional CTLS. Our results support its use as a screening tool. With the tube modulation technique, further adjustments in weighting IR and FBP algorithms based on body mass index become unnecessary.

Keywords: Computed tomography; Diagnostic reliability; Imaging quality iterative reconstruction; Lumbar spine; Patient safety; Radiation.

Publication types

  • Randomized Controlled Trial
  • Research Support, Non-U.S. Gov't

MeSH terms

  • Aged
  • Aged, 80 and over
  • Female
  • Humans
  • Lumbar Vertebrae / diagnostic imaging*
  • Male
  • Middle Aged
  • Prospective Studies
  • Radiation Dosage
  • Radiographic Image Interpretation, Computer-Assisted / methods*
  • Reproducibility of Results
  • Signal-To-Noise Ratio
  • Spinal Diseases / diagnostic imaging*
  • Tomography, X-Ray Computed / methods*