Purpose: To develop and evaluate a method based on ultrashort echo-time radial magnetic resonance (MR) imaging to quantify bone water (BW) concentration as a new metric of bone quality in human cortical bone in vivo.
Materials and methods: Human subject studies were institutional review board approved and HIPAA compliant; informed consent was obtained. Cortical BW concentration was determined with custom-designed MR imaging sequences at 3.0 T and was validated in sheep and human cortical bone by using exchange of native water with deuterium oxide (D(2)O). The submillisecond T2* of BW requires correction for relaxation losses during the radiofrequency pulse. BW was measured at the tibial midshaft in healthy pre- and postmenopausal women (mean age, 34.6 and 69.4 years, respectively; n = 5 in each group) and in patients receiving maintenance hemodialysis (mean age, 51.8 years; n = 6) and was compared with bone mineral density (BMD) at the same site at peripheral quantitative computed tomography, as well as with BMD of the lumbar spine and hip at dual x-ray absorptiometry. Data were analyzed by using the Pearson correlation coefficient and two-sided t tests as appropriate.
Results: Excellent agreement was obtained ex vivo between the water displaced by using D(2)O exchange and water measured with respect to a reference sample (r(2) = 0.99, P < .001). In vivo, BW in the postmenopausal group was greater by 65% (28.7% +/- 1.3 [standard deviation] vs 17.4% +/- 2.2, P < .001) than in the premenopausal group, and patients with renal osteodystrophy had higher BW (41.4% +/- 9.6) than the premenopausal group by 135% (P < .001) and the postmenopausal group by 43% (P = .02). BMD showed an opposite behavior, with much smaller group differences. Because the majority of BW is in the pore system of cortical bone, this parameter provides a surrogate measure for cortical porosity.
Conclusion: A new MR imaging-based method for quantifying BW noninvasively has been demonstrated.
RSNA, 2008