Low-dose MDCT: evaluation of the impact of systematic tube current reduction and sparse sampling on the detection of degenerative spine diseases

Objectives To investigate potential radiation dose reduction for multi-detector computed tomography (MDCT) exams of the spine by using sparse sampling and virtually lowered tube currents combined with statistical iterative reconstruction (SIR). Methods MDCT data of 26 patients (68.9 +/- 11.7 years, 42.3% males) were retrospectively simulated as if the scans were acquired at 50%, 10%, 5%, and 3% of the original X-ray tube current or number of projections, using SIR for image reconstructions. Two readers performed qualitative image evaluation considering overall image quality, artifacts, and contrast and determined the number and type of degenerative changes. Scoring was compared between readers and virtual low-dose and sparse-sampled MDCT, respectively. Results Image quality and contrast decreased with virtual lowering of tube current and sparse sampling, but all degenerative changes were correctly detected in MDCT with 50% of tube current as well as MDCT with 50% of projections. Sparse-sampled MDCT with only 10% of initial projections still enabled correct identification of all degenerative changes, in contrast to MDCT with virtual tube current reduction by 90% where non-calcified disc herniations were frequently missed (R1: 23.1%, R2: 21.2% non-diagnosed herniations). The average volumetric CT dose index (CTDIvol) was 1.4 mGy for MDCT with 10% of initial projections, compared with 13.8 mGy for standard-dose imaging. Conclusions MDCT with 50% of original tube current or projections using SIR still allowed for accurate diagnosis of degenerative changes. Sparse sampling may be more promising for further radiation dose reductions since no degenerative changes were missed with 10% of initial projections.

Automated summary

The study found that radiation dose for MDCT exams of the spine can be reduced by using sparse sampling and virtually lowered tube currents combined with statistical iterative reconstruction. However, image quality and contrast decreased with virtual lowering of tube current and sparse sampling.