Proton magnetic relaxation in bone marrow related to age and bone mineral density: low-resolution in vitro studies L. Lendinara a, *, C. Accorsi a , C. Agostini a , G. Angelini a , F. Baruffaldi b , M. Fini c , M. Motta a , G. Giavaresi c a Dipartimento di Fisica dell’Universita `, Bologna, Italy b Laboratorio di Tecnologia Medica - I.O.R. Bologna, Italy c Laboratorio di Chirurgia sperimentale - I.O.R. Bologna, Italy Received 12 December 1999; accepted 24 March 2001 Abstract Detailed analysis of proton spin-spin and spin-lattice relaxation behaviors of the bone marrow in the presence of trabecular bone network was performed at low-resolution ( B 0 = 0.496T) on rat vertebrae specimens deprived of spinal cord. Two groups of samples, from young and old healthy animals, were investigated before cellular necrosis had started. BMD measurements were carried out to quantify the expected age-related modifications of the trabecular bone network. 1 H-MR measurements were also performed on the same samples, deprived of marrow and saturated with water, in order to control the validity of a possible interpretation of the marrow 1 H-MR characteristics, in terms of marrow components, and to investigate the possible employment of these samples to study the trabecular bone network properties. We pointed out that: 1) a bimodal distribution of T 2i and T 1i values (distinguishing “fast” and “slow” relaxations) describes satisfactorily all the 1 H-MR experimental decays; 2) age-related modifications of the trabecular bone network are marked by correlate variations of the BMD value and of the proton spin-spin relaxation rates in water saturated samples; 3) age-related modifications of marrow are underlined by variations of the average value of the “fast” T 2i and of the “slow” T 1i relaxation time distributions, which could be attributed to the marrow components different from the fat granules of the adipose cells. Our results suggest that studies in vitro on bone tissue, by 1 H-MR techniques at low-resolution, may contribute to a better bone function characterization and, therefore, to a better clinical utilization of MRI techniques. © 2001 Elsevier Science Inc. All rights reserved. Keywords: Proton magnetic relaxometry; BMD measurements; Rat lumbar vertebrae; Age induced variations 1. Introduction With the progressive aging of the population, the pathol- ogies connected to osteoporosis phenomena are becoming a relevant voice in the national health services [1–3]. Osteo- porosis is characterized histologically by a quantitative and qualitative deficiency and by microarchitecture alterations of the osseous tissue and is known as a pathological stage of the progressive age-related deterioration of bone tissue. Different methods can be used, both radiological and not, to measure bone quality and to predict the associated fracture risk [4,5]. Dual-energy X-ray Absorptiometry (DXA) is a well known and widely accepted radiological approach to mea- sure Bone Mineral Density (BMD) [6 –11]. However, stud- ies have shown that changes in bone quality and structure, independent of bone mineral density, influence both bone strength and individual fracture risk [12,13]. Various techniques have been aimed at quantifying tra- becular bone structure in addition to bone density. Magnetic Resonance Imaging (MRI) has emerged as a promising method for the study of trabecular bone, due to its capability to discriminate elements of different magnetic susceptibil- ity, such as bone and bone marrow [14]. MRI is applied to image the bone marrow (high-signal intensity) surrounding the trabecular bone network (low-signal intensity). The ap- plication of MRI to assess bone structure has made a tre- mendous progress in the past few years [15–18] and several studies have shown that structural parameters derived from high-resolution MRI correlate significantly with trabecular BMD in vivo [19], demonstrating also age-dependency and fracture association [20]. * Corresponding author. Tel.: +51-2095137; fax: +51-2095047. E-mail address: lendinara@df.unibo.it (L. Lendinara). Magnetic Resonance Imaging 19 (2001) 745–753 0730-725X/01/$ – see front matter © 2001 Elsevier Science Inc. All rights reserved. PII: S0730-725X(01)00371-X