This article is more than 5 years old. Information may no longer be current.
Periostin levels higher in postmenopausal women with osteoporotic fracture
Click Here to Manage Email Alerts
Plasma periostin levels may be a biomarker for osteoporotic fracture risk, particularly for fractures in limbs and other non-spinal sites, according to research in Bone.
In a case-control study comparing postmenopausal Korean women with a history of non-vertebral and vertebral fractures with women without a history of fracture, researchers found that plasma periostin levels were inversely associated with proximal femur bone mineral density, but not with lumbar spine BMD.
Beom-Jun Kim, MD, PhD, of the division of endocrinology and metabolism at Asan Medical Center, University of Ulsan College of Medicine in Seoul, and colleagues analyzed data from 532 postmenopausal women who attended the osteoporosis clinic of Asan Medical Center between January 2011 and June 2012. Researchers identified 133 participants as having either non-vertebral (n = 81) or vertebral fractures (n = 62) within the study group; researchers assigned controls from the remaining 399 participants matched according to age (within 2.5 years) and BMI. Participants underwent BMD measurements with DXA at the lumbar spine and proximal femur; researchers obtained blood samples to measure plasma periostin concentrations and also measured serum calcium, serum phosphorus, glomerular filtration rate and bone turnover markers.
Before adjustments for BMD, parental history of fractures and other confounders, participants who had either non-vertebral or vertebral fractures had 11% higher plasma periostin levels when compared with controls (32.4 ng/mL vs. 29.2 ng/mL). After adjustment, participants with fractures had 12% higher periostin levels.
“Importantly, when total BMD was added as a confounding variable in this model, statistical significance persisted (10.4% higher in the case group),” the researchers wrote.
After separately analyzing participants with non-vertebral and vertebral fractures, researchers found that, before adjustment, women with non-vertebral fractures had 14% higher plasma periostin levels (32.6 ng/mL) than women with vertebral fractures (28.6 ng/mL); the rate was 12.8% higher after adjustment for confounders, including total femur BMD.
Within the highest periostin-concentration tertile, the adjusted odds ratio for osteoporotic fracture was 2.26-fold higher compared with the lowest tertile. For non-vertebral fractures, odds were 2.48-fold higher for women in the highest periostin tertile compared with women in the lowest tertile (95% CI, 1.10-5.61).
Researchers saw no association between plasma periostin and vertebral fracture after adjustment for confounders; there was also no association between plasma periostin and bone turnover markers.
Researchers also found that plasma periostin levels were inversely associated with total femur BMD both before and after adjustment, but not associated with lumber spine BMD.
“Importantly, the associations of plasma periostin with fracture risk and BMD were only observed in non-spinal skeletal sites, such as the limbs, rather than the spine,” the researchers wrote. “This association could be explained by the dominant expression of periostin in the periosteum, which covers mainly the long bones. These observations together suggest that periostin could serve as a biomarker for fracture risk, but that this may be dependent on the type of bone. Further interventional studies are needed to confirm this possibility.” – by Regina Schaffer
Disclosure: The researchers report no relevant financial disclosures.
Perspective
Back to Top
Bo Abrahamsen, MD, PhD
The findings reported by Kim and colleagues indicate that plasma periostin could serve as a marker of cortical bone turnover that may add to the prediction of non-vertebral osteoporotic fractures. They confirm prior findings by Rousseau and colleagues (J Clin Endocrinol Metab. 2014) that postmenopausal women with high periostin levels have twice the fracture risk of women with lower plasma levels of periostin and are also able to demonstrate a slightly lower femur bone mineral density with increasing plasma periostin levels. However, the strong association is with non-vertebral fracture, and this relationship is not attenuated by adjustment for BMD or for conventional bone turnover markers. While the intrinsic action of periostin appears to be to stimulate cortical bone formation, it seems plausible that plasma levels could serve as a marker of cortical bone turnover. This is very interesting because it has also been reported by Anastasilakis and colleagues (Horm Metab Res. 2014) that circulating periostin levels do not decline following infusion of zoledronic acid, a potent inhibitor of bone turnover. More work directly comparing turnover in cortical bone tissue with circulating periostin levels may provide insights into the mechanics of this novel bone marker.
Click Here to Manage Email Alerts