VOL. 83-B, NO. 1, JANUARY 2001 139 M. Fini, MD, Permanent Assistant in General Surgery G. Giavaresi, MD, Permanent Assistant in General Surgery P. Torricelli, BSc, Permanent Assistant in Biological Science Experimental Surgery Department, Research Institute Codivilla-Putti, Riz- zoli Orthopaedic Institute, 1/10 Via di Barbiano, 40136 Bologna, Italy. A. Krajewski, BChem, Senior Research Worker A. Ravaglioli, BChem, Research Director Institute for Technological Research on Ceramics of CNR-IRTEC, 64 Via Granarolo, 48018 Faenza, Ravenna, Italy. M. M. Belmonte, BSc, Research Worker G. Biagini, MD, Professor CIBAD Centre for Innovative Biomaterials, Institute of Normal Human Morphology, University of Ancona, 10/A Via Tronto, 60020 Ancona, Italy. R. Giardino, MD, Professor Department of Surgical Pathophysiology, Medical School of the Uni- versity of Bologna, 1/10 Via di Barbiano, 40136 Bologna, Italy. Correspondence should be sent to Dr M. Fini. ©2001 British Editorial Society of Bone and Joint Surgery 0301-620X/01/110162 $2.00 Biocompatibility and osseointegration in osteoporotic bone A PRELIMINARY IN VITRO AND IN VIVO STUDY M. Fini, G. Giavaresi, P. Torricelli, A. Krajewski, A. Ravaglioli, M. M. Belmonte, G. Biagini, R. Giardino From the Rizzoli Orthopaedic Institute, Bologna, Italy W e implanted nails made of titanium (Ti6Al4V) and of two types of glass ceramic material (RKKP and AP40) into healthy and osteopenic rats. After two months, a histomorphometric analysis was performed and the affinity index calculated. In addition, osteoblasts from normal and osteopenic bone were cultured and the biomaterials were evaluated in vitro. In normal bone the rate of osseointegration was similar for all materials tested (p < 0.5) while in osteopenic bone AP40 did not osseointegrate (p < 0.0005). In vitro, no differences were observed for all biomaterials when cultured in normal bone-derived cells whereas in osteopenic-bone-derived cells there was a significant difference in some of the tested parameters when using AP40. Our findings suggest that osteopenic models may be used in vivo in the preclinical evaluation of orthopaedic biomaterials. We suggest that primary cell cultures from pathological models could be used as an experimental model in vitro. J Bone Joint Surg [Br] 2001;83-B:139-43. Received 29 April 1999; Accepted after revision 13 September 1999 Biomaterials for use in orthopaedic surgery are usually evaluated by means of in vitro models using continuous or primary osteoblasts and primary differentiating osteogenic cell cultures, 1-6 and in vivo studies in which histomorpho- metric, ultrastructural and biomechanical evaluations allow the attachment of bone to the bone-material interface to be quantified. 7-9 In order to improve our preclinical knowledge of the biocompatibility and osseointegration properties of ortho- paedic implants under development, in vitro and in vivo, experimental models should not only be reliable but also be allied to the clinical situation. It should take into account that most implants are used in patients with altered bone mineralisation caused by osteoporosis 10,11 and that bone rarefaction is considered to be a negative predictive factor of osseointegration of bone implants and prostheses. 12,13 Hence, cells derived from osteopenic bone and osteopenic animal models should be used for the complete evaluation of biomaterials. 14 However, to our knowledge, biomaterials have not previously been tested on osteopenic bone-derived cells. We have therefore performed a preliminary study in vitro and in vivo using titanium (Ti6Al4V) and two biological glass ceramics (RKKP and AP40) on healthy and osteopen- ic rats. Our aim was to evaluate biomaterials in normal and osteopenic-bone-derived cells (in vitro study) and in nor- mal and osteopenic rats (in vivo study) and to compare the results. Materials and Methods For the in vivo study we used 36 cylindrical nails (2 mm in diameter and 3 mm in length) made of RKKP, AP40 and Ti6Al4V. AP40 and RKKP are silicophosphates of Ca, K and Na with similar composition and differ only in regard to the presence of amphoteric network-formers, La 3+ and Ta 5+ in RKKP, which contribute to the stabilisation of their molecular network. They were prepared by melting the starting products in platinum crucibles at 1450°C for 60 minutes. Glass nails were manufactured by casting the synthesised melted products from the crucible into a cylin- drical graphite die. In order to allow for glass thermal shrinkage the diameter of the die was slightly larger than the required final diameter of the nail. The height was set to the desired value by cutting and polishing.