DENVER – Computed tomography (CT) scans that were taken for an unrelated purpose could potentially be used to screen for osteoporosis, according to a new study. Researchers analyzed data from CT scans that produced estimates of bone mineral density (BMD) and femoral strength, and these performed similarly to dual-energy X-ray absorptiometry (DXA) in predicting fracture risk.
“The neat thing is that there’s no additional burden to the patients, because they’ve already had the CT scan. There’s no additional radiation exposure, no additional trip to the office. Another advantage to this is there are so many more men who are getting CT scans than there are who are getting DXAs for osteoporosis, so it’s an opportunity to screen more men,” Annette Adams, PhD, research scientist at Kaiser Permanente of Southern California, said in an interview at the annual meeting of the American Society for Bone and Mineral Research.
iStock/Thinkstock
To test the potential of the CT scans, the researchers conducted a case-control cohort analysis of patients aged 65 and over who were seen at 11 Kaiser Permanente Southern California (KPSC) hospitals. The patients had undergone abdominal or pelvic CT between 2006 and 2014. They had not experienced a fragility hip fracture before the CT scan was taken, but they had to have undergone a DXA within 3 years of the scan.
A total of 1,340 women and 619 men had a first hip fracture during the study period. They were compared to randomly selected subjects without hip fractures.
The researchers found associations between hip fractures and CT-based scores, and the relationships were stronger than those seen with DXA. In women, for each decrease in one standard deviation (SD) in hip BMD T-score, the hazard ratio (HR) for hip fracture was 2.18 (95% confidence interval [CI], 1.87-2.54). In men, the HR was 3.12 (2.35-4.14). The HRs for hip fracture based on DXA hip BMD T-score values were 1.80 (95% CI, 1.53-2.13) for women and 2.74 (95% CI, 2.15-3.49) for men.
CT-derived femoral strength values also performed well. In women, each one SD decrease in femoral strength seen in the CT-based scores was associated with an HR of 2.76 (95% CI, 2.25-3.39). In men, the value was HR 2.84 (95% CI, 2.20-3.66).
In a subanalysis of subjects who had not received osteoporosis treatment, for each one SD decrease in hip BMD T-score, the HR for hip fracture was 2.72 (95% CI, 2.24-3.32) in women and 3.93 (95% CI, 2.46-6.26) in men.
In untreated patients, each increase of one SD in femoral strength was tied to an HR in women of 3.81 (95% CI, 2.90-5.01) and 3.37 (95% CI, 2.27-5.01) in men.
The addition of established thresholds for osteoporosis (BMD T-score –2.5 or less) as well as fragile bone strength (3,000 Newtons or less in women, 3,500 Newtons or less in men) increased the 5-year sensitivity for hip fracture from 0.55 to 0.67 in women, and from 0.43 to 0.54 in men.
The technique is not yet ready for large-scale implementation because the process isn’t yet completely automated – it requires human review to eliminate some glitches, and that is likely to make it cost ineffective for now, Sally Warner, PhD, said in an interview. Dr. Warner is senior director of musculoskeletal imaging at Parexel, and was not involved in the study.
But she was nevertheless enthusiastic. “CT is such a great modality to be able to look at bone and interpret the density, the volume, the quality of the bone. It’s got great potential utility,” she said.
She said she thinks it will become cost-effective in time. “As long as the acquisition is standardized, I’m sure the automation could come a little bit more readily,” Dr. Warner said.
Amgen and Merck funded the study. Dr. Adams has received research funding from both companies. Dr. Warner reported having no financial disclosures.