On the Relationship Between Streaming Potential and Strain in an In vivo Bone Preparation B.R. Beck, 1 Y-X Qin, 2 K.J. McLeod, 2 and M.W. Otter 2 1 Grith University, School of Physiotherapy and Exercise Science, Private Mail Bag 50, Gold Coast Mail Centre, Queensland, 9726, Australia 2 Musculo-Skeletal Research Laboratory, Program in Biomedical Engineering, HSC-T18 030, State University of New York, Stony Brook, NY 11794-8181, USA Received: 13 December 2000 / Accepted: 8 January 2002 / Online publication: 12 August 2002 Abstract. Transcortical streaming potentials were mea- suredateachoftwocortical-surfacesiteswithrespectto a reference electrode in the medullary canal, in the left ulnae of six live, adult 2 yr-old), male 18.2  1.4 kg domestic turkeys, under general anesthesia, for each of two loading conditions. We observed that the relation- ship among streaming potential magnitude, surface strain, and strain gradient is not as simple as anticipat- ed. Under predominantly axial and bending load con- ditions,signi®cantlydierentstrainandstraingradients were generated at the two recording sites. However, no signi®cant dierences were detectable in transcortical streaming potentials for one of the loading conditions, and only a slight dierence was detected in the other. Conversely, correlations of streaming potential magni- tude to strain at both sites show robust relationships r 2 = 0.45, P £ 0.02),albeitwithdierentslopesforthe two sites. These ®ndings may have implications for the contribution of streaming potentials, or at least, ¯uid ¯ow to the stimulation of bone cells. Key words: Streaming potentials ÐStrain ÐStrain gradients ÐBone electrokinetics ÐMechanical loading Ample evidence exists of the innate ability of bone to adapttochronicalterationsinmechanicalloading[1±8], however, the mechanism by which bone cells sense mechanical load information in order to initiate the adaptive response is uncertain. Establishment of the exactnatureofloadtransductionsignalswouldnotonly solve a fundamental mystery in the ®eld of bone re- search, but would provide a physiological basis for fu- ture development of bone pathology therapies. At the organ level, bone appears to respond to me- chanical loading in a dose-dependent manner [9]. That is, chronic reductions and increases within injurious limits) in mechanical loading precipitate net bone loss andgain,respectively,inproportiontothemagnitudeof the applied load. Such a direct relationship, however, is more dicult to establish at the tissue level. Indeed, the dynamic strain and strain gradient distribution within long bones during normal physiological loading is highly non-uniform and is non-uniformly dependent on gait speed [10±12]. While bone accretion is occasionally observed at the site of peak-induced strain following periods of chronic experimental loading [6, 7, 13], ele- vated bone formation is also likely to be observed at sitesofminimumengenderedstrain[3,5,9,14].Thatan increase in long-term dynamic strain can eect a gen- eralized increase in bone mass with minimal response fromboneatthesiteofpeakdeformation,suggeststhat bone-surface strain per se is not directly responsible for transduction of the load signal to cells. Instead, the adaptivesignalislikelytooriginatefromaphenomenon secondary to the strain event. Recently, there have been reports of correlations be- tween the spatial gradients of strain and the location of periosteal bone formation [15]. In vivo, bone strain gra- dients produce measurable electrokinetic potentials across the tissue [16±18]. The preponderance of recent evidence indicates that stress-generated potentials, that can be recorded from the surface of bone during me- chanical loading, are a product of streaming potentials generated by ¯uid ¯ow through spaces in the bone ma- trix[16,18±21,23].Itispossiblethatsuchalterationsin the electrical environment are perceived by bone cells andinitiateachainofintracellularevents,whichleadto adaptive bone resorption and/or formation [24±28]. While there is considerable evidence that transcortical streaming potential magnitudes are locally correlated to strain and strain gradient magnitudes, it is not clear whetherstreamingpotentialsmeasuredatdierentbone sites will directly re¯ect the local bone matrix-strain en- vironment. Variation with position may re¯ect, for ex- ample,ageometricvariationinzetapotential[21,29]or the superposition of ¯uid ¯ow from some other source such as intramedullary IM) pressurization [30, 31]. Thepurposeofthecurrentstudywastocomparethe magnitude of transcortical streaming potentials at sites Calcif Tissue Int 2002) 71:335±343 DOI: 10.1007/s002230020050 Calci®ed Tissue International Ó 2002 Springer-Verlag New York Inc. Correspondence to: B. R. Beck