Feasibility analysis of low-dose CT with asynchronous quantitative computed tomography to assess vBMD

Tingting Hu, Xingyuan Yang, Lei Gao, Ying Liu, Wei Zhang, Yan Wang, Xiaona Zhu, Xiangdong Liu, Hongran Liu, Xiaohui MaBMC Medical Imaging 2023 · DOI

TL;DR

This study investigates the feasibility of using low-dose CT (LDCT) scans, commonly used for lung cancer screening, to also assess volumetric bone mineral density (vBMD) using an asynchronous quantitative CT (asynchronous QCT) method. The findings suggest that LDCT with asynchronous QCT can accurately measure vBMD without additional radiation exposure, patient time, or cost, making it a valuable opportunistic screening tool.

What Problem Does This Paper Address?

Assessing bone mineral density (BMD) is crucial for diagnosing and managing osteoporosis, a condition characterized by low bone mass and increased fracture risk, particularly in postmenopausal women. Traditional quantitative CT (QCT) methods for measuring volumetric BMD (vBMD) typically require a calibration phantom placed on the patient during the scan. However, this adds to the radiation dose and scan time. Asynchronous QCT offers a way to assess vBMD without an in-scan phantom, relying instead on regular quality assurance scans. While conventional-dose CT (CDCT) has been used with asynchronous QCT, the use of low-dose CT (LDCT) for this purpose, especially in the context of opportunistic screening during lung cancer evaluations, has been less explored. This paper aims to bridge this gap by evaluating the feasibility and accuracy of LDCT with asynchronous QCT for vBMD assessment, comparing it to CDCT.

Methods

The researchers conducted a retrospective study involving 416 women aged 26-45. Participants underwent chest CT examinations that included both asynchronous QCT and either an LDCT or CDCT scanning protocol, which they self-determined. The study excluded individuals with conditions known to affect bone metabolism.

Key methodological aspects include:

  • Asynchronous QCT: This technique was employed to measure vBMD in the T12-L2 vertebral bodies. It relies on a phantom that is scanned regularly for quality assurance, rather than a phantom placed with the patient during each scan.
  • LDCT vs. CDCT Categorization: Participants were categorized into LDCT and CDCT groups based on their effective dose (ED), with a threshold of 1.0 mSv used to differentiate.
  • Image Acquisition and Analysis: Chest CT images were transferred to a QCT workstation for vBMD analysis. Elliptical regions of interest were automatically placed within the trabecular bone of the T12-L2 vertebrae, avoiding cortical bone.
  • Statistical Analysis: Standard statistical methods, including ANOVA, independent samples t-tests, and Pearson/Spearman correlations, were used to compare vBMD across age groups, between LDCT and CDCT groups, and to assess correlations with clinical and imaging parameters (age, BMI, mAs, CTDIvol, DLP).

Datasets

The study utilized a dataset of 416 women patients who underwent CT examinations at a single center between January 2020 and April 2022. No new benchmarks or datasets were introduced; the study relied on existing patient data.

Key Insights and Ideas

  • Opportunistic Screening Potential: The most significant insight is that LDCT, already a common tool for lung cancer screening, can be leveraged to assess vBMD simultaneously without additional radiation, patient time, or cost. This transforms a routine diagnostic scan into a dual-purpose screening tool.
  • No Significant vBMD Difference: A crucial finding is that vBMD measurements were comparable between the LDCT and CDCT groups, despite differences in radiation dose parameters (mAs, CTDIvol, DLP). This suggests LDCT is sufficiently sensitive for vBMD assessment.
  • vBMD Independence from Radiation Dose Parameters: The study found no correlation between vBMD and radiation dose parameters (mAs, CTDIvol, DLP) in either the LDCT or CDCT groups. This is important because it implies that variations in radiation dose within these protocols do not compromise the accuracy of vBMD measurement.
  • Age and BMI Not Correlated with vBMD in this Cohort: Interestingly, in this specific cohort of premenopausal women (aged 25-45), vBMD was not correlated with age or BMI. This contrasts with findings in older populations where age-related bone loss is more pronounced.

Key Results

  • There was no statistically significant difference in vBMD between the LDCT and CDCT groups (P ≥ 0.480).
  • vBMD was not correlated with age, BMI, mAs, CTDIvol, or DLP in either the LDCT or CDCT groups.
  • Significant differences were observed in mAs, CTDIvol, and DLP between the LDCT and CDCT groups, as expected due to the different dose protocols.
  • BMI was correlated with mAs, CTDIvol, and DLP in both groups, indicating that higher BMI is associated with higher radiation dose parameters.
  • mAs was strongly correlated with CTDIvol and DLP in both groups, confirming the expected relationship between these parameters.

Limitations and Future Work

The authors acknowledge several limitations:

  • Limited Age Range and Gender: The study only included women aged 26-45. Future work should include larger, multicenter studies with both male and female participants across a wider age range to validate these findings and assess vBMD distribution differences.
  • Self-Determined Dose Protocols: Participants self-determined their LDCT or CDCT protocol, which could introduce bias.
  • Automated Tube Current Modulation: The use of automated tube current modulation rather than fixed mAs, while beneficial for image quality across different body sizes, adds a variable.
  • Single Radiologist Analysis: vBMD was measured by only one radiologist, limiting reproducibility data.

Future research should focus on expanding the study population and investigating the long-term implications of using LDCT with asynchronous QCT for vBMD assessment in clinical practice.

Summary updated

BibTeX
@article{hu2023feasibility,
  author = {Tingting Hu and Xingyuan Yang and Lei Gao and Ying Liu and Wei Zhang and Yan Wang and Xiaona Zhu and Xiangdong Liu and Hongran Liu and Xiaohui Ma},
  title = {{Feasibility analysis of low-dose CT with asynchronous quantitative computed tomography to assess vBMD}},
  journal = {BMC Medical Imaging},
  year = {2023},
  doi = {10.1186/s12880-023-01115-1},
}
Figure