Association between low-frequency ultrasound and hip fractures - comparison with DXA-based BMD

This is an elec onic ep in o he o iginal a icle. This ep in may di e om he o iginal in pagina ion and ypog aphic de ail. Au ho (s): Ti le: Yea : Ve sion: Please ci e he o iginal e sion: All ma e ial supplied ia JYX is p o ec ed by copy igh and o he in ellec ual p ope y igh s, and duplica ion o sale o all o pa o any o he eposi o y collec ions is no pe mi ed, excep ha ma e ial may be duplica ed by you o you esea ch use o educa ional pu poses in elec onic o p in o m. You mus ob ain pe mission o any o he use. Elec onic o p in copies may no be o e ed, whe he o sale o o he wise o anyone who is no an au ho ised use . Associa ion be ween low- equency ul asound and hip ac u es - compa ison wi h DXA-based BMD Mää ä, Mikko; Moilanen, Pe o; Timonen, Jussi; Pulkkinen, Pasi; Ko pelainen, Raija; Jämsä, Timo Mää ä, M., Moilanen, P., Timonen, J., Pulkkinen, P., Ko pelainen, R., & Jämsä, T. (2014). Associa ion be ween low- equency ul asound and hip ac u es - compa ison wi h DXA-based BMD. BMC Musculoskele al Diso de s, 15(208). h ps://doi.o g/10.1186/1471-2474-15-208 2014 RESEARCH ARTICLE Open Access Associa ion be ween low- equency ul asound and hip ac u es −compa ison wi h DXA-based BMD Mikko Mää ä 1,2* , Pe o Moilanen 3 , Jussi Timonen 3 , Pasi Pulkkinen 1 , Raija Ko pelainen 1,4,5,6 and Timo Jämsä 1,6,7 Abs ac Backg ound: New me hods o diagnosing os eopo osis and e alua ing ac u e isk a e being de eloped. We aim o s udy he associa ion be ween low- equency (LF) axial ansmission ul asound and hip ac u e isk in a popula ion-based sample o olde women. Me hods: The s udy popula ion consis ed o 490 communi y-dwelling women (78–82 yea s). Ul asound eloci y (V LF ) a mid- ibia was measu ed in 2006 using a low- equency scanning axial ansmission de ice. Bone mine al densi y (BMD) a p oximal emu measu ed using dual-ene gy x- ay abso p iome y (DXA) was used as he e e ence me hod. The ac u e his o y o he pa icipan s was collec ed om Decembe 1997 un il he end o 2010. Li es yle- ela ed isk ac o s and mobili y we e assessed a 1997. Resul s: Du ing he o al ollow-up pe iod (1997–2010), 130 women had one o mo e ac u es, and 20 o hem had a hip ac u e. Low V LF ( he lowes qua ile) was associa ed wi h inc eased hip ac u e isk when compa ed wi h V LF in he no mal ange (Odds a io, OR = 3.3, 95% con idence in e al (CI) 1.3-8.4). Howe e , V LF was no ela ed o ac u e isk when all bone si es we e conside ed. Os eopo o ic emo al neck BMD was associa ed wi h highe isk o a hip ac u e (OR = 4.1, 95% CI 1.6-10.5) and highe isk o any ac u e (OR = 2.4, 95% CI 1.6-3.8) compa ed o he non-os eopo o ic emo al neck BMD. Dec eased V LF emained a signi ican isk ac o o hip ac u e when combined wi h li es yle- ela ed isk ac o s (OR = 3.3, 95% CI 1.2-9.0). Conclusion: Low V LF was associa ed wi h hip ac u e isk in olde women e en when combined wi h li es yle- ela ed isk ac o s. Fu he de elopmen o he me hod is needed o imp o e he measu emen p ecision and o con i m he esul s. Keywo ds: Quan i a i e ul asound, Speed o sound, Os eopo osis, Hip ac u e, Co ical bone Backg ound Os eopo o ic ac u es possess a signi ican public heal h p oblem ha is inc easing due o aging popula ion. A he momen , he golden s anda d used in ac u e isk assessmen is bone mine al densi y (BMD) measu emen s using dual ene gy x- ay abso p iome y (DXA). Recen ly, Wo ld Heal h O ganiza ion (WHO) in oduced FRAX®, a ac u e isk calcula o ha combines easily ob ained clin- ical in o ma ion and DXA-based emo al neck BMD, i a ailable, o es ima e he 10-yea os eopo o ic ac u e p obabili y [1]. Quan i a i e ul asound (QUS) has aised in e es as an al e na i e me hod o x- ay-based imaging o meas- u ing bone s a us. The e is a numbe o di e en QUS me hods ha ha e been used o assess bone s a us and o e alua e ac u e isk [2]. The ad an ages o ul asound include a ela i ely low cos and a po able echnique wi h no ionizing adia ion. The esul s o QUS measu emen s ha e been shown o be associa ed wi h ac u e isk [3]. A ques ion has a isen whe he QUS pa ame e s can se e as a su oga e o DXA-based BMD o imp o e FRAX® es ima e [4]. * Co espondence: [email p o ec ed] 1 Depa men o Medical Technology, Uni e si y o Oulu, Ins i u e o Biomedicine, PO Box 5000, FI-90014 Oulu, Finland 2 In o ech Oulu, Uni e si y o Oulu, Oulu, Finland Full lis o au ho in o ma ion is a ailable a he end o he a icle © 2014 Mää ä e al.; licensee BioMed Cen al L d. This is an Open Access a icle dis ibu ed unde he e ms o he C ea i e Commons A ibu ion License (h p://c ea i ecommons.o g/licenses/by/2.0), which pe mi s un es ic ed use, dis ibu ion, and ep oduc ion in any medium, p o ided he o iginal wo k is p ope ly c edi ed. Mää ä e al. BMC Musculoskele al Diso de s 2014, 15:208 h p://www.biomedcen al.com/1471-2474/15/208 A no el low- equency (LF) axial ansmission QUS me hod o assessing he p ope ies o co ical bone has been shown o e lec bone densi y and co ical hick- ness [5-7]. A ela ionship be ween he LF ibial eloci y (V LF ) and geome y o p oximal emu in olde women has also been sugges ed [8]. Howe e , he sui abili y o he me hod o indi idual ac u e isk assessmen has no ye been es ablished. In his s udy we e alua ed he associa ion be ween he LF axial ansmission ul asound me hod and ac u es in a popula ion-based sample o olde women, using he s anda d DXA as he e e ence me hod. We hypo he- sized ha dec eased ul asound eloci y measu ed on ibia is associa ed wi h inc eased hip ac u e isk in eld- e ly emales. In ou p e ious popula ion based p ospec i e s udy we assessed li e-s yle ela ed de e minan s o hip ac u e in elde ly emales [9]. He e we analyzed how he combina ion o low- equency ul asound and p e iously de e mined li es yle- ela ed ac o s a e ela ed o he isk o hip ac u e. Me hods Subjec s and clinical assessmen The s udy popula ion consis ed o 490 women bo n be- ween 1924 and 1927, o iginally ec ui ed in 1997 as a popula ion-based coho o s udy he isk ac o s o os eopo osis and ac u es [10]. All hose s ill ali e belong- ing o he o iginal coho we e in i ed o clinical measu e- men s in 2006. A o al o 618 women a ended hese measu emen s, and 490 we e measu ed using quan i a i e ul asound. The ac u e his o y be ween Decembe 1, 1997 and Decembe 31, 2010 was collec ed om hospi al discha ge egis e s. The ac u es we e con i med manu- ally om medical eco ds o a oid he bias o eco ding mul iple hospi aliza ions due o a single ac u e. Heal h and li es yle in o ma ion, including medical his o y, age a menopause, smoking habi s, alcohol and co ee consump- ion, physical ac i i y and mobili y, calcium and i amin D in akes, and ac u e his o y was collec ed a s udy base- line in 1997 using sel -adminis a i e ques ionnai es and in e iews. The assessmen o unc ional mobili y a base- line was done using he “Timed Up & Go”(TUG) es [11]. All subjec s ga e a w i en in o med consen and he s udy p o ocol was app o ed by he E hics Commi ee o he No he n Os obo hnia Hospi al Dis ic . The s udy was done in acco dance wi h he Decla a ion o Helsinki. Quan i a i e ul asound measu emen s The speed o sound was eco ded in he medial mid-sha o he le ibia using a scanning low- equency (LF) axial ansmission de ice. The p inciple o he de ice and he measu emen se up has been published p e iously [7,8]. In b ie , wo sepa a e ansduce s ( c = 200 kHz) moun ed on a ail scanned a 30 mm dis ance. The ime o ligh o he i s a i ing signal (FAS) was de e mined using he i s maximum. The appa en eloci y o he FAS (V LF ) was de e mined by measu ing he ime o ligh o a numbe o sou ce- ecei e dis ances. The p ecision e o was cha ac e ized by CV ms which was 3.2%. The s an- da dized coe icien o a ia ion (SCV) [12] o he me hod was 6.6%. Due o he lack o es ablished e e ence popula- ion da a, we used he p esen s udy popula ion o de ine he ange o SCV. DXA measu emen s A Hologic DXA de ice (Delphi QDR se ies, Hologic, Bed o d, MA, USA) was used o ge he e e ence da a. S anda d an e opos e io posi ioning was used o meas- u e he emo al neck bone mine al densi y (BMD) o he le p oximal emu . S a is ical analysis The subjec s o he s udy we e di ided in o h ee g oups based on hei ac u e his o y (Figu e 1): a) women wi h- ou ac u es (NF, con ols), b) women wi h any ac u e (Fx), and c) women wi h a hip ac u e (Hip Fx). All he hip ac u e pa ien s we e also included in g oup (b). The ac u e da a o h ee di e en ime pe iods we e analyzed: a) om he beginning o he ollow-up pe iod un il he ime o bone measu emen s (1997–2006), b) om he ime o he bone measu emen s un il he end o he ollow-up pe iod (2006–2010), and c) he whole ollow-up pe iod (1997–2010). The women wi h ac u es we e compa ed o hose wi hou ac u es. Since he da a we e no mally dis ibu ed he independen samples - es was used o analyze he s a is ical signi icance o he di e - ences be ween he NF, Fx, and Hip Fx g oups. The s udy subjec s we e also classi ied in o no mal, os eopenic, and os eopo o ic acco ding o hei emo al neck BMD T- sco e in acco dance wi h he WHO de ini ion [13]. Due o he lack o an es ablished e e ence popula ion, T-sco es could no be calcula ed o low- equency ul asound el- oci y (V LF ). Thus, V LF esul s we e di ided in o h ee g oups based on qua iles wi hin he s udy popula ion: a) low V LF (0-25%), b) mode a e V LF (25-50% and 50-75% combined), and c) high V LF (75-100%). C oss abula ion and he χ 2 es we e used o compa e he dis ibu ion o he subjec s wi h ac u es and hose wi hou ac u es wi hin hese g oups. To u he analyze he s a is ical sig- ni icance o he associa ion o bone measu emen esul s (QUS/DXA) wi h he ac u e isk, a mul i a ia e logis ic eg ession analysis was used. The esul s o hose analyses a e epo ed as he odds a ios (OR) and 95% con idence in e als (CI). The imaging modali y (QUS/DXA) used was included in he model and he o wa d s epwise (like- lihood a io) me hod was used o o m he inal models. All models we e adjus ed wi h age and BMI. Fo he ac- u es ha occu ed a e he measu emen s in 2006, he Mää ä e al. BMC Musculoskele al Diso de s 2014, 15:208 Page 2 o 8 h p://www.biomedcen al.com/1471-2474/15/208 p edic i e abili y o each bone measu emen alue was assessed using he Cox p opo ional haza ds model and he co esponding haza d a ios (HR), and he 95% CIs we e calcula ed. A simila p o ocol o logis ic eg ession wasused.Inaddi ion,p e ious ac u es(1997–2006) (yes, no) was added o models as co a ia es. The ollow- up ime be ween he measu emen ime and he ime o he i s ac u e, dea h, o end o he ollow-up pe iod was eco ded. Logis ic eg ession models we e also calcula ed o analyze he e ec o combining V LF and p e iously de e mined [9] li es yle- ela ed isk ac o s o hip ac u e. In ou p e ious popula ion based co- ho s udy BMI, unc ional mobili y, physical ac i i y, hype ension, co ee consump ion, and daily smoking p edic ed hip ac u es in 1222 old women. These ac- o s we e used as a co a ia es o de e mine he bes i model o he cu en subpopula ion wi h ul asound and DXA measu emen s. Figu e 1 Conso cha . Numbe o women wi h and wi hou ac u es a di e en imepoin s. Mää ä e al. BMC Musculoskele al Diso de s 2014, 15:208 Page 3 o 8 h p://www.biomedcen al.com/1471-2474/15/208 Table 1 Dis ibu ion o ac u es (n (%)) acco ding o he ICD-10 code wi h selec ed subdi isions in a popula ion-based sample o olde women (n = 490) 1997–2006 2006–2010 1997–2010 M80 Os eopo osis wi h pa hological ac u e 2 (2.6) 2 (1.1) S12 F ac u e o neck 1 (1.3) 1 (0.6) S22 F ac u e o ib(s), s e num and ho acic spine 3 (3.1) 3 (3.9) 6 (3.4) S32 F ac u e o lumba spine and pel is S32.0 F ac u e o lumba e eb a 9 (11.8) 9 (5.2) O he 3 (3.1) 4 (5.3) 7 (4.0) S42 F ac u e o shoulde and uppe a m 11 (11.2) 6 (7.9) 17 (9.8) S52 F ac u e o o ea m S52.5 F ac u e o lowe end o adius 29 (29.6) 22 (28.9) 51 (29.3) O he 5 (5.1) 1 (1.3) 6 (3.4) S62 F ac u e a w is and hand le el 7 (7.1) 5 (6.6) 12 (6.9) S72 F ac u e o emu S72.0 F ac u e o neck o emu 4 (4.1) 7 (9.2) 11 (6.3) S72.1 Pe ochan e ic ac u e 1 (1.0) 4 (5.3) 5 (2.9) S72.2 Sub ochan e ic ac u e 4 (4.1) 4 (2.3) O he 3 (3.1) 6 (7.9) 9 (5.2) S82 F ac u e o lowe leg, including ankle 26 (26.5) 5 (6.6) 31 (17.8) S92 F ac u e o oo , excep ankle 2 (2.0) 1 (1.3) 3 (1.7) To al numbe o ac u es 98 (100) 76 (100) 174 (100) ICD-10 = In e na ional Classi ica ion o Diseases, 10 h e ision. Time pe iods ep esen p e ious ac u es (1997–2006) and u u e ac u es (2006–2010) wi h espec o he bone measu emen s, and all ac u es du ing he ollow-up pe iod (1997–2010). Table 2 Cha ac e is ics o he women wi h di e en ac u e s a us (NF, Fx, and Hip Fx) in di e en ime pe iods (1997–2006, 2006–2010, and 1997–2010) a he ime o he measu emen s (2006) (n = 490) NF Fx 1997-2010 Hip Fx 1997-2010 n =360 n = 130 n = 20 Fx 1997-2006 Fx 2006-2010 Hip Fx 1997-2006 Hip Fx 2006-2010 n=81 n=61 n=9 n=11 Age [yea s] 79.9 (1.2) 80.0 (1.2) 80.0 (1.3) 79.7 (1.2) 80.3 (1.1) * 79.3 (1.4) 80.5 (0.9) Weigh [kg] 67.7 (11.3) 67.8 (12.2) 69.8 (15.7) 69.5 (12.3) 65.8 (12.2) 75.7 (10.2) * 65.0 (18.2) Heigh [cm] 155.2 (5.4) 155.6 (5.8) 158.9 (7.2) * 156.6 (5.4) * 154.5 (6.4) 160.7 (4.4) * 157.4 (8.8) BMI [kg/m 2 ] 28.1 (4.5) 28.0 (4.7) 27.4 (4.6) 28.4 (4.8) 27.5 (4.6) 29.3 (3.6) 25.9 (4.9) V LF [m/s] 3583 (193) 3547 (200) 3515 (245) 3534 (204) * 3564 (203) 3500 (291) 3527 (215) Femo al neck BMD [g/cm 2 ] 0.654 (0.102) 0.606 (0.083) * 0.589 (0.102) * 0.609 (0.078) * 0.605 (0.096) * 0.611 (0.089) 0.570 (0.112) * Values a e Mean (SD). * Independen samples - es p- alue < 0.05 when compa ed o NF g oup. Mää ä e al. BMC Musculoskele al Diso de s 2014, 15:208 Page 4 o 8 h p://www.biomedcen al.com/1471-2474/15/208 S a is ical analyses we e done using SPSS o Win- dows (Release 18.0, SPSS Inc., Chicago, IL, USA). In all es s, p- alues less han 0.05 we e conside ed s a is i- cally signi ican . Resul s Du ing he i s ollow-up pe iod (1997–2006), 98 ac- u es occu ed in 81 women (Figu e 1, Table 1). Be ween 2006 and 2010, 61 subjec s sus ained ac u es; in o al 76 di e en ac u es occu ed. Al oge he , be ween he yea s 1997 and 2010, 174 di e en ac u es occu ed in 130 subjec s. W is ac u e was he mos common ac- u e (n = 51). The e we e 9 hip ac u es be ween 1997 and 2006, and 11 hip ac u es a e he 2006 measu e- men s, o aling in 20 hip ac u es du ing he whole ollow-up pe iod. Thi y- i e subjec s died du ing he ollow-up pe iod. The cha ac e is ics o di e en ac u e g oups a e shown in Table 2. The women wi h a hip ac u e we e alle han he women wi hou ac u es (p < 0.05). Those wi h a p e- ious ac u e (1997–2006) had a lowe V LF han he ones wi hou ac u es (p < 0.05). Subjec s who sus ained a ac u e du ing he ollow-up pe iod had a lowe emo al neck BMD han hose in he NF g oup (p < 0.05), excep when compa ing he Hip Fx 1997–2006 and NF g oups. Di e ences be ween dis ibu ions in he NF and Hip Fx g oups we e obse ed in bo h V LF and emo al neck BMD (p < 0.05) (Table 3). Also, he e was a di e ence in he dis ibu ion o subjec s based on emo al neck BMD be ween he Fx and NF g oups. Based on he eg ession analysis, low V LF was no as- socia ed wi h p e ious ac u es (1997–2006) o wi h he ac u es ha occu ed a e he measu emen s (2006– 2010) (Table 4). Dec eased emo al neck BMD, howe e , was associa ed wi h inc eased ac u e isk in all ollow- up pe iods; es ima ed isks (odds and haza d a ios) a - ied be ween 1.8 and 2.5 compa ed o he women wi h emo al neck T-sco e > −2.5. Dec eased V LF was associa ed wi h an inc eased isk o hip ac u e occu ed be o e he measu emen s (1997– 2006) (OR = 6.3; Table 5). Addi ionally, ac oss he whole ollow-up pe iod (1997–2010), low V LF was associa ed wi h highe isk o hip ac u e compa ed o mode a e o high V LF (OR = 3.3). An os eopo o ic emo al neck BMD p edic ed hip ac u es a e he measu emen s (2006–2010) wi h a haza d a io (HR) o 4.8 (95% CI 1.4-16.6) o subjec s wi h a T-sco e ≤−2.5 compa ed o he women wi h no mal o os eopenic T-sco e. Low emo al neck BMD was also associa ed wi h an inc eased isk o hip ac u e (OR = 4.1, 95% CI 1.6-10.5) du ing he whole ollow-up pe iod (1997–2010) compa ed o women wi h no mal o os eopenic emo al neck BMD. When V LF was combined wi h p e iously assessed [9] hip ac u e isk ac o s eg ession model o he same popula ion (Table 6), he inal model included V LF , low physical ac i i y, and impai ed unc ional mobili y (p = 0.001, Table 6), V LF being he s onges independen ac- o o hip ac u e (OR = 3.3, 95% CI 1.2–9.0). Femo al neck BMD did no each s a is ical signi icance when in- cluded in he eg ession analyses along wi h li es yle- ela ed isk ac o s. Table 3 The dis ibu ion o women wi h and wi hou a ac u e and hip ac u e acco ding o he V LF and BMD alues NF Fx 1997–2010 Hip Fx 1997–2010 (n = 360) (n = 130) (n = 20) V LF Highes 25% 101 (28) 24 (18) 5 (25) * 25% o 75% 170 (47) 69 (53) 5 (25) Lowes 25% 89 (25) 37 (28) 10 (50) Femo al neck BMD No mal 73 (20) 11 (8) * 2 (10) * Os eopenic 206 (57) 65 (50) 8 (40) Os eopo o ic 81 (23) 54 (42) 10 (50) Values a e n (%). * Dis ibu ion o he g oup is di e en han ha o he NF g oup (p- alue o χ 2 es < 0.05). Table 4 Associa ion o low V LF and os eopo o ic emo al neck BMD wi h ac u es ha occu ed du ing di e en ollow-up pe iods Fx 1997-2006 Fx 2006-2010 Fx 1997-2010 NF n = 409, Fx n = 81 NF n = 429, Fx n = 61 NF n = 334, Fx n = 121* n OR (95% CI) p HR (95% CI) p OR (95% CI) p V LF Mode a e o high (25-100%) 364 Low (0-25%) 126 1.6 (0.9 - 2.6) 0.088 0.8 (0.4 - 1.4) 0.359 1.3 (0.8 - 2.0) 0.326 Femo al neck BMD T-sco e > −2.5 355 T-sco e ≤−2.5 135 1.8 (1.1 - 3.0) 0.019 2.4 (1.4 - 3.9) 0.001 2.5 (1.6 - 3.9) <0.001 All eg ession models we e adjus ed by age and BMI. In Cox eg ession p e ious ac u e (du ing 1997–2006) was also used as a co a ia e. NF = non- ac u ed g oup, Fx = ac u ed g oup, n = numbe o subjec s, OR = odds a io o logis ic eg ession, HR = haza d a io o Cox eg ession, CI = con idence in e al, p = p- alue o he co a ia e in he model. *Subjec s who died du ing he ollow-up we e excluded (NF n = 26, Fx n = 9). Mää ä e al. BMC Musculoskele al Diso de s 2014, 15:208 Page 5 o 8 h p://www.biomedcen al.com/1471-2474/15/208 Discussion In his popula ion-based s udy, we used a low- equency ul asound scanne o assess he associa ion be ween dec eased ul asound eloci y and ac u e isk. To ou knowledge, his is he i s ime ha low- equency axial ansmission ul asound is being used in a popula ion- based coho o olde women. The epo ed esul s show ha dec eased ul asound eloci y (V LF ) was associa ed wi h highe isk o hip ac u e compa ed o mode a e o high V LF . DXA and LF ul asound yielded simila esul s when compa ing he hip ac u e and non- ac u e g oups. Ou indings a e in line wi h ea lie s udies on he ac u e-disc imina ion abili y o simila axial ansmis- sion ul asound de ices. Using he Omnisense (Sunligh , BeamMed L d, Pe ah Tik a, Is ael) de ice, ope a ing a a cen e equency o 1.25 MHz, Nguyen e al. [14] epo ed ha dec eased ibial SOS is associa ed wi h inc eased ac- u e isk independen ly o BMD and age wi h an odds a- io (OR) o 1.75. The co esponding OR as de e mined o he emo al neck BMD he e was 2.11. Damilakis e al. [15] ound no di e ence using he Omnisense de ice in he ibial SOS o heal hy subjec s and subjec s wi h os eopo - o ic ac u e. Ne e heless, hey epo ed inc eased os eo- po o ic ac u e ORs o ul asound measu emen s on he adius and phalanx (ORs be ween 1.7 and 2.7). In o he s udies wi h an Omnisense de ice, SOS measu ed on he adius has been shown o disc imina e subjec s wi h hip ac u e om con ols wi h no ac u es (ORs a ying be- ween 1.9 and 2.7) [16-18]. Ano he axial ansmission QUS de ice, My iad Soundscan (My iad Ul asound sys- ems, Is ael) uses a cen e equency o 250 kHz which is close o ha used in his s udy. Using his de ice, S egman e al. [19] epo ed a low-ene gy appendicula ac u e OR o 1.4 o he ibial SOS. This obse a ion is in good ag eemen wi h he p esen s udy, whe e we ound a simila end in he e ospec i e pa o he s udy. Auga e al. [20] ound an inc eased OR o 1.7 o he ib- ial SOS o subjec s who sus ained hip ac u e compa ed o non- ac u e con ols. The co esponding OR o he emo al neck BMD was 3.5 in hei s udy. This SOS OR is somewha lowe han hose obse ed in he p esen s udy o hip ac u es. Also, Talman e al. [21] s udied he ac- u e disc imina ion abili y o a 1 MHz bidi ec ional ul a- sonome e . They epo ed a ac u e OR o 1.8 o he adius SOS and o 2.1 o he emo al neck BMD. Di ec compa ison be ween he ul asound me hods used is di icul due o di e ences in he ul asound me hod- ology and measu emen se up. Also, he bone si e and ype o ac u e, as well as he s udy popula ions, a ec he esul s. Howe e , he majo i y o he s udies imply ha axial ansmission ul asound is capable o asses- sing ac u e isk. Some s udies ha e assessed he combining clinical isk ac o s and QUS pa ame e s [22-24]. In he cu en s udy low V LF emained a signi ican isk ac o o hip ac u e when included in he li es yle- ela ed isk ac o model ob- se ed in ou p e ious s udy wi h he same popula ion [9]. Howe e , he li es yle- ela ed isk ac o s we e collec ed a baseline whe eas QUS measu emen s we e made a eigh Table 5 Associa ion o low V LF and os eopo o ic emo al neck BMD wi h hip ac u es ha occu ed du ing di e en ollow-up pe iods Hip Fx 1997-2006 Hip Fx 2006-2010 Hip Fx 1997-2010 NF n = 409, Hip Fx n = 9 NF n = 429, Hip Fx n = 11 NF n = 334, Hip Fx n = 19* n OR (95% CI) p n HR (95% CI) p n OR (95% CI) p V LF Mode a e o high (25-100%) 313 324 260 Low (0-25%) 105 6.3 (1.5 - 25.5) 0.010 116 1.7 (0.5 - 5.7) 0.415 93 3.3 (1.3 - 8.4) 0.012 Femo al neck BMD T-sco e > −2.5 311 325 267 T-sco e ≤−2.5 107 1.5 (0.4 - 6.0) 0.593 115 4.8 (1.4 - 16.6) 0.013 86 4.1 (1.6 –10.5) 0.003 All eg ession models we e adjus ed by age and BMI. In Cox eg ession p e ious ac u e (du ing 1997–2006) was also used as a co a ia e. NF = non- ac u ed g oup, Hip Fx = hip ac u e g oup, n = numbe o subjec s, OR = odds a io o logis ic eg ession, HR = haza d a io o Cox eg ession, CI = con idence in e al, p = p- alue o he co a ia e in he model. *Subjec s who died du ing he ollow-up we e excluded (NF n = 26, Hip Fx n = 1). Table 6 Logis ic eg ession models wi hou and wi h V LF o ha ing a hip ac u e in a popula ion-based sample o olde women OR (95% CI) p- alue Li es yle- ela ed isk ac o s 1 0.002 * TUG ≥11 s s. less ( e e en ) 3.4 (1.2 - 9.9) 0.026 Low PA s. mode a e o high ( e e en ) 2.8 (1.0 - 7.5) 0.046 Co ee consump ion > 3 cups/day s. less ( e e en ) 0.3 (0.1 - 1.0) 0.051 Li es yle- ela ed isk ac o s and V LF2 0.001 * Low V LF (0-25%) s. Mode a e o high (25-100%) ( e e en ) 3.3 (1.2 - 9.0) 0.018 Low PA s. mode a e o high ( e e en ) 3.1 (1.1 - 8.5) 0.028 TUG ≥11 s s. less ( e e en ) 3.1 (1.0 - 8.9) 0.042 Odds a ios (OR) a e calcula ed compa ed o he NF g oup. TUG “Timed Up & Go” es , PA physical ac i i y, CI con idence in e al. The numbe o subjec s in he analyses was Hip Fx n = 18, NF n = 296. 1 Age, BMI, TUG, PA, hype ension, co ee consump ion, and smoking we e included in he analysis o o m he bes model, 2 The bes model a e including V LF in he analysis, * p- alue o he ull model. Mää ä e al. BMC Musculoskele al Diso de s 2014, 15:208 Page 6 o 8 h p://www.biomedcen al.com/1471-2474/15/208 yea s om baseline. Thus, he esul s need o be con- i med in u u e p ospec i e s udies. In he p esen s udy ul asound eloci y was measu ed a mid- ibia. E en hough os eopo osis is a sys emic disease, he deg ee o changes in bone s uc u e a ies be ween bone si es. In weigh -bea ing si es (e.g. he ibia) bone loss is no as big as in non-weigh -bea ing si es (e.g. he adius and phalanx) [25]. Thus, e en i he ibia is an easily ac- cessible si e o axial ul asound measu emen s, i may no be hemos esponsi esi e oos eopo o icchanges.Re- cen ly, we ha e epo ed ha LF axial- ansmission ul a- sound in he adius is able o e ospec i ely disc imina e pos menopausal women wi h ac u es om age-ma ched con ols wi h no ac u es using an imp o ed e sion o he LF ul asound de ice used he e [26]. The s eng h o he p esen s udy was he use o a popula ion-based coho o olde women. The o he s eng h was he ela i ely long ollow-up ime be o e and a e he bone measu emen s. Howe e , his s udy also had some limi a ions. One majo limi a ion was he limi ed measu emen p ecision o he cu en p o o ype de ice. The bulky mechanical scanning se up limi s he posi ioning accu acy and does no enable p ope co ec- ion o he impac s o o e lying so issue [8]. As a e- sul , he in i o p ecision (CV ms )waslimi ed.Du ing he ollow-up pe iod o he s udy, signi ican mechanical imp o emen s we e implemen ed in he de ice, and a CV o 0.5% could be achie ed wi h he la es de ice e - sion [5,26]. Ob iously, his aluable in o ma ion a ailable a e he de ice imp o emen canno e ospec i ely im- p o e he quali y o he esul s p esen ed he e. The o he limi a ion is he e ospec i e na u e o he s udy. I is possible ha he ac u e e en s be o e he measu emen s caused he changes in bone p ope ies (e.g. ia al e ed loading condi ions) and a ec ed on ou come o he mea- su emen . Also, he numbe o hip ac u es was ela i ely low, which signi ican ly limi ed he s a is ical eliabili y o he esul s. Addi ionally, a mino limi a ion was he lack o e e ence da a o young and heal hy popula ion, which disabled he de e mina ion o he T-sco e. Ins ead, we used qua iles o de ine he subjec s wi h low V LF alues. Conclusions In conclusion, dec eased low- equency ul asound e- loci y was associa ed o inc eased hip ac u e isk despi e he limi ed measu emen p ecision. The esul s epo ed he e can be used o u he imp o e he measu emen p ecision o he me hod so as o eliably p edic u u e ac u es. To his end ca e ully planned ollow-up s udies a e needed. Compe ing in e es s The au ho s decla e ha hey ha e no compe ing in e es s. Au ho s’con ibu ions MM pa icipa ed in he ac u e da a collec ion, ca ied ou he main analyses, and d a ed he manusc ip . PM pa icipa ed in he analyses and in e p e a ion o ul asound da a and manusc ip w i ing. JT and PP made a con ibu ion o he design o he s udy and e ising he manusc ip . RK and TJ pa icipa ed in s udy concep ion and design, da a in e p e a ion and made subs an ial con ibu ions o he manusc ip . RK also pa icipa ed signi ican ly o collec ion o li es yle- ela ed isk ac o s da a. All au ho s ha e ead, e ised, and gi en hei inal app o al o he e sion o be published. Acknowledgemen s The wo k done is a pa o he Os eopo o ic F ac u e Risk s udy suppo ed by he Finnish Funding Agency o Technology and Inno a ion (Tekes, p ojec s no. 40463/05, 40403/06). The wo k was also suppo ed by he Academy o Finland. MM was suppo ed by Tauno Tönning Founda ion and Finnish Cul u al Founda ion. Au ho de ails 1 Depa men o Medical Technology, Uni e si y o Oulu, Ins i u e o Biomedicine, PO Box 5000, FI-90014 Oulu, Finland. 2 In o ech Oulu, Uni e si y o Oulu, Oulu, Finland. 3 Depa men o Physics, Uni e si y o Jy äskylä, Jy äskylä, Finland. 4 Depa men o Spo s and Exe cise Medicine, Oulu Deaconess Ins i u e, Oulu, Finland. 5 Ins i u e o Heal h Sciences, Uni e si y o Oulu, Oulu, Finland. 6 Medical Resea ch Cen e Oulu, Oulu Uni e si y Hospi al and Uni e si y o Oulu, Oulu, Finland. 7 Depa men o Diagnos ic Radiology, Oulu Uni e si y Hospi al, Oulu, Finland. Recei ed: 28 Augus 2013 Accep ed: 10 June 2014 Published: 16 June 2014 Re e ences 1. Kanis JA, Oden A, Johansson H, Bo gs öm F, S öm O, McCloskey E: FRAX and i s applica ions o clinical p ac ice. Bone 2009, 44(5):734–743. 2. Glüe CC: A new quali y o bone ul asound esea ch. IEEE T ans Ul ason Fe oelec F eq Con ol 2008, 55(7):1524–1528. 3. Ma ín F, González-Macías J, Díez-Pé ez A, Palma S, Delgado-Rod íguez M: Rela ionship be ween bone quan i a i e ul asound and ac u es: a me a-analysis. J Bone Mine Res 2006, 21(7):1126–1135. 4. Lewiecki EM, Comps on JE, Mille PD, Adachi JD, Adams JE, Leslie WD, Kanis JA, Moayye i A, Adle RA, Hans DB, Kendle DL, Díez-Pé ez A, K ieg MA, Mas i BK, Lo enc RR, Baue DC, Blake GM, Josse RG, Cla k P, Khan AA, FRAX® Posi ion De elopmen Con e ence Membe s: O icial Posi ions o FRAX® Bone Mine al Densi y and FRAX® simpli ica ion om Join O icial Posi ions De elopmen Con e ence o he In e na ional Socie y o Clinical Densi ome y and In e na ional Os eopo osis Founda ion on FRAX®. J Clin Densi om 2011, 14(3):226–236. 5. Kilappa V, Moilanen P, Xu L, Nicholson PH, Timonen J, Cheng S: Low- equency axial ul asound eloci y co ela es wi h bone mine al densi y and co ical hickness in he adius and ibia in p e- and pos menopausal women. Os eopo os In 2011, 22(4):1103–1113. 6. Moilanen P, Nicholson PH, Kä kkäinen T, Wang Q, Timonen J, Cheng S: Assessmen o he ibia using ul asonic guided wa es in pube al gi ls. Os eopo os In 2003, 14(12):1020–1027. 7. Nicholson PH, Moilanen P, Kä kkäinen T, Timonen J, Cheng S: Guided ul asonic wa es in long bones: modelling, expe imen and in i o applica ion. Physiol Meas 2002, 23(4):755–768. 8. Mää ä M, Moilanen P, Nicholson P, Cheng S, Timonen J, Jämsä T: Co ela ion o Tibial Low-F equency Ul asound Veloci y wi h Femo al Radiog aphic Measu emen s and BMD in Elde ly Women. Ul asound Med Biol 2009, 35(6):903–911. 9. Mää ä M, Te ho E, Jokinen H, Pulkkinen P, Ko pelainen J, Heikkinen J, Keinänen-Kiukaanniemi S, Jämsä T, Ko pelainen R: Li es yle ac o s and si e-speci ic isk o hip ac u e in communi y dwelling olde women – a 13-yea p ospec i e popula ion-based coho s udy. BMC Musculoskele Diso d 2012, 13(1):173. 10. Ko pelainen R, Ko pelainen J, Heikkinen J, Väänänen K, Keinänen-Kiukaanniemi S: Li es yle ac o s a e associa ed wi h os eopo osis in lean women bu no in no mal and o e weigh women: a popula ion-based coho s udy o 1222 women. Os eopo os In 2003, 14(1):34–43. 11. Podsiadlo D, Richa dson S: The imed “Up & Go”: a es o basic unc ional mobili y o ail elde ly pe sons. J Am Ge ia Soc 1991, 39(2):142–148. Mää ä e al. BMC Musculoskele al Diso de s 2014, 15:208 Page 7 o 8 h p://www.biomedcen al.com/1471-2474/15/208 12. O gee JM, Fos e H, McCloskey EV, Khan S, Coombes G, Kanis JA: A p ecise me hod o he assessmen o ibial ul asound eloci y. Os eopo os In 1996, 6(1):1–7. 13. Repo o a WHO S udy G oup: Assessmen o ac u e isk and i s applica ion o sc eening o pos menopausal os eopo osis. Wo ld Heal h O gan Tech Rep Se 1994, 843:1–129. 14. Nguyen TV, Cen e JR, Eisman JA: Bone mine al densi y-independen associa ion o quan i a i e ul asound measu emen s and ac u e isk in women. Os eopo os In 2004, 15(12):942–947. 15. Damilakis J, Papadokos akis G, V aho i i H, Tsaga aki I, Pe isinakis K, Hadjipa lou A, Gou soyiannis N: Ul asound eloci y h ough he co ex o phalanges, adius, and ibia in no mal and os eopo o ic pos menopausal women using a new mul isi e quan i a i e ul asound de ice. In es Radiol 2003, 38(4):207–211. 16. Hans D, S i as a SK, Singal C, Ba kmann R, Njeh CF, Kan o o ich E, Glüe CC, Genan HK: Does combining he esul s om mul iple bone si es measu ed by a new quan i a i e ul asound de ice imp o e disc imina ion o hip ac u e? J Bone Mine Res 1999, 14(4):644–651. 17. Hans D, Gen on L, Allaoua S, Picha d C, Slosman DO: Hip ac u e disc imina ion s udy: QUS o he adius and he calcaneum. J Clin Densi om 2003, 6(2):163–172. 18. Weiss M, Ben-Shlomo A, Hagag P, Ish-Shalom S: Disc imina ion o p oximal hip ac u e by quan i a i e ul asound measu emen a he adius. Os eopo os In 2000, 11(5):411–416. 19. S egman MR, Heaney RP, T a e s-Gus a son D, Leis J: Co ical ul asound eloci y as an indica o o bone s a us. Os eopo os In 1995, 5(5):349–353. 20. Auga P, Fan B, Lane NE, Lang TF, LeHi P, Lu Y, U mann M, Genan HK: Assessmen o bone mine al a appendicula si es in emales wi h ac u es o he p oximal emu . Bone 1998, 22(4):395–402. 21. Talman M, Kol a S, Roux C, Haguenaue D, Vedel I, Cassou B, Bossy E, Laugie P: In i o pe o mance e alua ion o bi-di ec ional ul asonic axial ansmission o co ical bone assessmen . Ul asound Med Biol 2009, 35(6):912–919. 22. Du osie C, Hans D, K ieg MA, Ru ieux C, Co nuz J, Meunie PJ, Scho AM: Combining clinical ac o s and quan i a i e ul asound imp o es he de ec ion o women bo h a low and high isk o hip ac u e. Os eopo os In 2007, 18(12):1651–1659. 23. Ka jalainen JP, Riekkinen O, Töy äs J, Hakulinen M, K öge H, Rikkonen T, Salo aa a K, Ju elin JS: Mul i-si e bone ul asound measu emen s in elde ly women wi h and wi hou p e ious hip ac u es. Os eopo os In 2012, 23(4):1287–1295. 24. Hans D, Du osie C, Kanis JA, Johansson H, Scho -Pe helaz AM, K ieg MA: Assessmen o he 10-yea p obabili y o os eopo o ic hip ac u e combining clinical isk ac o s and heel bone ul asound: he EPISEM p ospec i e coho o 12,958 elde ly women. J Bone Mine Res 2008, 23(7):1045–1051. 25. Knapp KM, Blake GM, Spec o TD, Fogelman I: Mul isi e quan i a i e ul asound: p ecision, age- and menopause- ela ed changes, ac u e disc imina ion, and T-sco e equi alence wi h dual-ene gy X- ay abso p iome y. Os eopo os In 2001, 12(6):456–464. 26. Moilanen P, Mää ä M, Kilappa V, Xu L, Nicholson PH, Alén M, Timonen J, Jämsä T, Cheng S: Disc imina ion o ac u es by low- equency axial ansmission ul asound in pos menopausal emales. Os eopo os In 2012, 24(2):723–730. doi:10.1186/1471-2474-15-208 Ci e his a icle as: Mää ä e al.:Associa ion be ween low- equency ul asound and hip ac u es −compa ison wi h DXA-based BMD. BMC Musculoskele al Diso de s 2014 15:208. Submi you nex manusc ip o BioMed Cen al and ake ull ad an age o : • Con enien online submission • Tho ough pee e iew • No space cons ain s o colo figu e cha ges • Immedia e publica ion on accep ance • Inclusion in PubMed, CAS, Scopus and Google Schola • Resea ch which is eely a ailable o edis ibu ion Submi you manusc ip a www.biomedcen al.com/submi Mää ä e al. BMC Musculoskele al Diso de s 2014, 15:208 Page 8 o 8 h p://www.biomedcen al.com/1471-2474/15/208