- Email: [email protected]
Force platform for rats measures fore and hind forces concurrently
ARTICLE IN PRESS Journal of Biomechanics 42 (2009) 2734–2738
Contents lists available at ScienceDirect
Journal of Biomechanics journal homepage: www.elsevier.com/locate/jbiomech www.JBiomech.com
Force platform for rats measures fore and hind forces concurrently Jo M. Welch a,b,, Jeremie A. Wade c, Ben M. Hillberry c, Connie M. Weaver a a b c
Foods and Nutrition, Purdue University, West Lafayette, IN, USA Health and Human Performance, Dalhousie University, Halifax, NS, Canada B3H 3J5 Mechanical Engineering, Purdue University, West Lafayette, IN, USA
a r t i c l e in f o
a b s t r a c t
Article history: Accepted 11 August 2009
Animal models are commonly used to test the efficacy of impact loading regimens on bone strength. We designed an inexpensive force platform to concurrently measure the separate peak vertical impact forces produced by the fore and hindfeet of immature F-344 rats when dropped onto the platform. The force platform consisted of three load cells placed in a triangular pattern under a flat plate. Rats were dropped from heights of 30, 45 and 60 cm onto the platform so that they landed on all four feet concurrently. The peak vertical impact forces produced by the feet of the rats were measured using a sampling frequency of 100 kHz. The location of each foot at landing relative to the load cells, and the force received by each load cell were combined in a series of static equations to solve for the vertical impact forces produced by the fore and hindfeet. The forces produced by feet when rats stood on the single platform were similarly determined. The forces exerted separately by the fore and hindfeet of young rats when landing on the plate as a ratio to standing forces were then calculated. Rats when standing bore more weight on their hindfeet but landed with more weight on their forefeet, which provides rationale for the greater response to landing forces of bones in the forelimbs than those in the hindlimbs. This system provided a useful method to simultaneously measure peak vertical impact forces in fore and hindfeet in rats. & 2009 Elsevier Ltd. All rights reserved.
Keywords: Force platform Ground reaction forces Rat Bone Impact exercise
1. Introduction The minimum ground reaction force (GRF) that is required to activate beneficial bone remodelling appears to differ with age, propensity for bone strength, and type of habitual exercise. Knowledge of which exercise protocols might trigger bone change during growth or aging, and under various other conditions could reduce rates of fracture in humans. Bone responds positively to forces that meet or exceed a threshold (Frost, 1987). High impact activities such as those delivered by the track and field sport of triple jump are highly osteogenic (Heinonen et al., 2001), while low impact activities, such as walking, produce little change in bone (Palombaro, 2005). Axial impact forces can improve long bone strength (Welch et al., 2004; Heinonen et al., 1996) and are typically measured using a force platform. Resultant force values are typically normalized using standing forces and reported as either a multiple of body weight (Quatman et al., 2006) or as the ratio of forces, N landing/N standing (Bassey and Ramsdale, 1995). The use of rodents as models to test exercise protocols for osteogenicity offers the advantages of precise control of dietary, Corresponding author at: School of Health and Human Performance, 6230 South Street, Dalhousie University, Halifax, Nova Scotia, Canada B3H 3J5. Tel.: + 1 902 494 2475; fax: + 1 902 494 5120. E-mail address: [email protected] (J.M. Welch).
0021-9290/$ - see front matter & 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.jbiomech.2009.08.002
exercise, and environmental conditions, the ability to measure bone strength directly, and large reductions in the experimental time needed to detect changes in bone. In animals, the effects on bone of impact from jumping, or from freefalling onto a surface, has been reported in rats (Umemura et al., 1997, 1995, 2002b; Honda et al., 2001, 2003b, 2003a; Nagasawa and Umemura, 2002; Welch et al., 2004), mice (Lee et al., 2002; Kodama et al., 2000; Umemura et al., 2002a), and poultry (Judex and Zernicke, 2000). Methods used to measure GRF in humans require modification for use in quadrupeds. Several methods to quantify GRFs during rodent locomotion have been reported. Howard and coworkers (Howard et al., 2000) walked adult rats in two directions through a narrow channel that allowed only ipsilateral limbs to pass over the force platform. Webb et al. (2003); Howard et al. (2000) trotted young rats over a raised runway with three small platforms embedded beneath it to independently capture the forces from each limb. Zumwalt et al. (2006) walked mice over a force plate that was modified so that only a small area was sensitive to applied forces. Commercially available small-force force plates systems, such as the HE6 6, (Advanced Mechanical Technology, Incorporated, Watertown, MA) that was used by Zumwalt et al. (2006) have maximum sampling frequencies that are too low to capture the peak vertical forces present at an impact from drop jumps and therefore cannot accurately capture
ARTICLE IN PRESS J.M. Welch et al. / Journal of Biomechanics 42 (2009) 2734–2738
the peak landing forces. Quantifying the ground reaction force received by one leg when landing from a height requires a different method. Methods commonly used for measuring these forces in humans include two parallel force platforms (Oggero et al., 1997; Quatman et al., 2006), or landing with one foot on the platform and the other on the floor beside the platform (Seegmiller and McCaw, 2003). Although humans bear similar weight on both feet when landing from drop jumps (Quatman et al., 2006), quadrupeds display different proportions of weight on their fore and hindfeet (Welch et al., 2004). Fore and hind differences are also specific to each strain of rat, and to the activity that they are performing (Webb et al., 2003). Rodents also differ from humans in that they cannot be expected to land oriented as the researcher prefers and most will only momentarily stand in the location where they landed. It is therefore important to be able to differentiate the peak force from subsequent ones and to differentiate between where the feet landed and subsequent steps. A successful force platform to assess vertical impact in rats therefore requires a high sampling frequency and an adjustable landing area. In earlier experiments we examined the effects of freefall landings on the geometry and tissue properties of leg bones in growing rats that were dropped from 30, 60 cm (Welch et al., 2004) and 45 cm (Welch et al., 2008). Results indicated substantial osteogenic benefits from this activity in the forelimb bones but muted or absent responses in the hindlimbs. We therefore hypothesized that the hindlimbs received lower impact forces than did the forelimbs. To test this hypothesis, we constructed a force platform that allowed concurrent measurement of the fore and hind peak vertical impact forces in rats at landing from a fixed drop height. Preliminary GRF data from this method were previously reported for the 30-, 60-cm (Welch et al., 2004) and 45-cm (Welch et al., 2008) drop distances. Presently, we are reporting the description of the force platform system used to determine the vertical ground reaction forces exerted on the fore and hindfeet of rats landing from heights of 30, 45, and 60 cm.
2735
platform when the rats landed from the different heights. Additionally, these standard weights were placed at other locations on the force plate and the values at the load cells compared favourably with calculated values. The calibration curves included zero and therefore the system was tested over the full range of values obtained during the experiment. When the rats were dropped, a solid boundary was placed around and nearly touching the force plate (not shown) as we observed that it improved the relaxation of the animals and facilitated their retention on the lab bench. Standard graph paper was attached to the top of the force platform so that the locations of the feet after the drop could be marked and these locations could then be recorded as coordinates relative to the load cell locations. The rats can be dropped in any orientation so long as the locations of the paw landings relative to the load cells are included in the equations.
2.3. Protocol Fischer-344 rats tend to be a nervous strain and therefore the rats were habituated to human contact and procedures over a three week period. Each rat was picked up daily by a handler who grasped its body with a bare hand, lifted, petted, and carried it about the room. After one week, rats began habituation to the dropping protocol. An initial three very low drops progressed in number, height, and increased frequency until the final regimen assigned to each rat was reached. Rats were randomized, based on body weight, to 30, 45, or 60 cm dropping heights (n= 4 each group). All force platform measurements for all rats were recorded within a 24-h period, with different rats selected at each drop height to avoid either changes in body size that would occur due to growth if the experiment took multiple days, or the physical stress from 30 instead of 10 landings per rat on one day. We previously observed that most trained rats remained relaxed, as demonstrated in inverted U-shaped position of the rat in Fig. 2, for at least 15 freefalls per session. On the data collection day, each rat was placed on an ink pad containing washable ink and then lifted until its feet reached the prescribed height. It was dropped so that it landed on all feet approximately concurrently. Non-simultaneous landings were audible on the aluminum platform; results from those drops were excluded and additional drops performed. The coordinates for the center of each footprint were measured from the ink prints on the graph paper (Fig. 3). The peak values of the forces recorded concurrently by each load cell at the
2. Materials and methods 2.1. Animals Twelve female Fischer-344 rats, aged 46 days, were purchased from Harlan Sprague–Dawley, Inc. (Indianapolis, IN, USA). This strain of inbred rats was the same strain, sex, and age as those used previously to test the efficacy of the freefall impact protocol on osteogenicity (Welch et al., 2004, 2008). This strain is more fragile-boned than others (Turner et al., 2001) and therefore is ideal for testing the effects of physical forces on bone because fragile-boned rodents are more responsive to such forces (Akhter et al., 1998). On the day the measurements were made, the rats were 67 days old, had a mean mass of 140 g, and were growing at a mean rate of 1.6 g d 1. Animal protocols and care were approved by the Purdue Animal Care and Use Committee of Purdue University.
Fig. 1. The three-load cell force platform: (A) force platform, (B) base plate, (C) load cell.
2.2. Apparatus A 17.78 cm 17.78 cm aluminium force platform was attached above three 2.27 kg load cells (P/N AL311AT; Honeywell Sensotec, Columbus, OH, USA). An 11.43 cm 11.43 cm steel base plate (2.0 kg) was attached to the bottom of the load cells. The force plate system is shown in Fig. 1. When assembling the system, the locations of the load cells form a triangle but their precise locations can vary, as can the relative locations of the load cells to each other. The load cells were connected to three analog channels via a 68-pin shielded I/O connector block (P/N SCB68, National Instruments, Austin, TX, USA). A shielded cable connected this breakout box to a multifunction DAQ card (P/N PCI-6034E, National Instruments). Data were sampled at 100 kHz by a custom program using Labview 7 Express (National Instruments). Before data collection, to test for linearity and accuracy, calibration of each load cell was performed while connected to the platform. Five standard weights of 1, 2, 3, 4 and 5 lb, corresponding to forces of approximately 4, 9, 13, 18, 22 N, were placed sequentially directly over each load cell and calibration curves were constructed from this data for each load cell. These calibration weights were selected because they spanned the range of forces received by the
Fig. 2. A rat in freefall flight. A relaxed rat travels with its legs and back forming an inverted U-shape, and appears to look for its landing surface.
ARTICLE IN PRESS 2736
J.M. Welch et al. / Journal of Biomechanics 42 (2009) 2734–2738 averaged and assumed to be the same for all rats. Standing forefeet weight for each rat was then calculated as follows: Standing forefeet weight ¼ rat mass g k where g = 9.81 m s 1, and k is the proportion of rat mass on the forefeet when standing. The peak vertical impact forces produced by fore and hindfeet when a rat was dropped onto the force platform were standardized by dividing each rat’s fore and hindfoot forces by body weight, respectively, when standing. Contralateral legs were assumed to bear equal force. 2.4. Statistical analysis Differences between groups were tested by ANOVA using SPSS version 14.0. Pairwise comparisons between means were made post hoc using the Tukey’s HSD test. An a level was set at po 0.05 for all tests. We hypothesized that rats could modify their impacts by landing from different heights with different foot stances. To address this, the distances between fore, hind, and fore to hindfeet at the moment of landing from the three drop heights were tested for difference using ANOVA with body mass as a covariate. Coefficients of variation (CV) for the peak landing forces and landing stances were calculated from the pooled variance of four individual variances per treatment; variances for individual rats were computed from the 10 replications per rat.
3. Results Fig. 3. Rat footprints on the landing grid. X and Y coordinates for each foot were determined from the corner of the platform nearest the left anterior of the rat. All rats were dropped to face the same direction.
Table 1 Nomenclature and equations used in calculation of fore and hindfeet forces. Nomenclature C F x y
Load cell peak force Foot peak vertical impact force Horizontal coordinate Vertical coordinate
Subscripts a–c 1 2 3 4
Load cells a, b, c Right forefoot Right hindfoot Left hindfoot Left forefoot
Equations (1)
xfore = 12(x1 + x4)
(2)
yfore = 12(y1 + y4)
(3)
xhind = 12(x2 +x3)
(4)
yhind = 12(y2 + y3) Ffore(xfore xa)+ F2(x2 xa)+ F3(x3 xa) = Cb(xb xa) +Cc(xc xa) Ffore(yfore ya) + F2(y2 ya) + F3(y3 ya) = Cb(yb ya) +Cc(yc ya) Ffore + F2 +F3 = Ca +Cb + Cc
(5) (6) (7)
moment of impact were entered into the custom program. Feet coordinates were combined with load cell maximal forces to calculate the peak vertical impact force using the following method. (Nomenclature and equations are defined in Table 1.) Spatial locations for the midpoint of the fore and hindfeet were calculated and denoted as xfore, yfore, xhind, and yhind (Eqs. (1–4)). F1, F2, F3, and F4 represented the vertical impact forces on each foot at landing. F1 and F4 represented the forefeet and F2 and F3 represented the hindfeet. To reduce four forces to three force unknowns to be used with three static equations, a force for the forefeet, Ffore, occurring at (xfore, yfore), was created. Three simultaneous equations (Eqs. (5–7)) were solved to obtain values for Ffore, F2, and F3, where the latter two variables are the GRFs of each hindfoot. The corresponding force for the hindfeet, Fhind, occurring at (xhind, yhind), was calculated by subtracting Ffore from the total impact forces. The proportion of mass borne by the fore and hindfeet of the rats when standing was measured by placing rats from each group on the force platform. Mean fore and hind values from a total of 10 repetitions using six rats were used to calculate the proportion of weight borne by the fore and hindfeet. Other repetitions were discarded because the animals would not stand still. The proportions of forces borne by the fore and hindfeet from this subset of rats were
The specially designed force plate with associated methodology and customized analysis was used to quantify the landing forces of rats dropped from 30, 45, and 60 cm. The forefeet of rats were measured as one force, and both hindfeet together as another, separate force. Standing rats bore an average of 41.6% of their weight on their forefeet. Rats dropped from 30, 45, and 60 cm bore 65.8%, 59.3%, and 60.7% of the impact forces on their forefeet, respectively. Peak vertical impact forces from 10 drops per rat were averaged and the mean forces for each rat are presented in Table 2. At landing, the forces borne by the feet relative to those borne by the same feet when standing were significantly greater in the fore than the hindfeet (Table 2, Fig. 4). In the forefeet each increment of drop height resulted in significantly greater landing forces. However, in the hindfeet, the forces borne at 45 and 60 cm were greater than at 30 cm, but were not significantly different from each other. Total landing force had a much lower variability between drops and between rats in the same groups than did fore and hind forces. Stance at landing did not differ between groups. The CV for the peak vertical landing force of the fore, hind, and total feet, and the landing stances, are presented in Table 3.
4. Discussion A method to measure peak vertical impact forces in the fore and hindfeet of rats dropped onto one force platform is described. This simple bench top system is useful in quantifying vertical ground reaction forces in rodents and is especially well suited to measure these forces in animals used as models for impact exercise research. We were able to determine that rats, when landing on four feet concurrently from a height, bore more weight on their forefeet than hindfeet. This knowledge was useful to explain the previous findings that increased strength in the forelimb bones in young rats that were dropped 10 times per day from 30 or 60 cm for eight weeks correlated with the greater forces borne by the front limbs (Welch et al., 2004). Peak impact forces increased linearly with drop height in the forelimbs but in the hindlimbs, forces approached an asymptote at the 45 cm height. Drop jumps by gymnasts have elicited a linear increase in vertical GRF with height of drop (McKay et al., 2005; Seegmiller and McCaw, 2003) through a range of low heights but this linearity was blunted at greater jump heights, which was
ARTICLE IN PRESS J.M. Welch et al. / Journal of Biomechanics 42 (2009) 2734–2738
2737
Table 2 Forces measured and ground reaction forces calculated from rat freefalls onto force platforma. Distance dropped
Landing F, all feet (N) Landing F, all feet/rat mass (N/g) Fraction landing F, front feet (%) Landing F, each front foot (N) Landing F, front foot/rat mass (N/g) Landing F, each hindfoot (N) Landing F, back foot/rat mass (N/g) Mass of rat (g) Fraction of standing F, calculated, front feet (%)b Front foot forces, landing/standingc Hindfoot forces, landing/standingd
30 cm
45 cm
60 cm
10.15 70.08 0.07570.001 65.80 71.13 3.34 70.08 0.025 70.000 1.74 70.05 0.013 70.000 136 73 41.56 12.04 70.24 4.46 70.20
14.15 70.55*** 0.099 70.001*** 59.29 70.62* 4.19 70.18* 0.029 70.001*** 2.88 70.10*** 0.020 70.001*** 143 74 41.56 14.43 70.32** 7.11 70.13***
15.917 0.35***y 0.1127 0.002***yyy 60.747 1.99 4.84 7 0.25*** 0.0347 0.001***yyy 3.117 0.11*** 0.0227 0.001*** 1427 3 41.56 16.747 0.52***yy 7.77 0.43***
Different from 30 cm: *p o 0.05, **p o0.01, and ***p o0.001. 60 cm is different from 45 cm: ypo 0.05, yyp o0.01, and yyyp o 0.001. a
Each data point represents the mean values from four rats7SE. Constant calculated from 10 repetitions with six rats and extrapolated to all rats. Standing weight or force on forefeet calculated from rat mass g k, where g= 9.81 m s 1, and k is the proportion of rat mass on the forefeet. d Standing weight or force on hindfeet calculated from rat mass g (1 k), where g = 9.81 m s 1, and k is the proportion of rat mass on the forefeet. b c
Peak impact force (N*N-1)
19
*** †
16
Table 3 Coefficients of variation (%) for the rat force platform.
**
Drop distance (cm) 30 (n= 4)
45 (n= 4)
60 (n= 4)
Peak vertical forces Total (four feet) Forefeet Hind feet
5.2 10.9 20.1
6.9 11.4 18.4
10.2 14.7 19.3
Landing stancea Fore left to right Hind left to right Mean fore to hind
11.9 20.0 14.3
14.6 13.2 9.7
17.5 17.2 8.9
13 10
***
7
***
4 1
a
0
30 45 Drop distance (cm)
Landing stance refers to the distances between feet at landing.
60
Fig. 4. Peak vertical impact forces produced by rats dropped from heights of 30, 45, and 60 cm. Data are means 7 SD; n= 4. Solid line represents forefoot impact force per forefoot weight; dashed line represents hindfoot impact force per hindfoot weight. For each line, statistical differences are denoted as follows: ** p o 0.01, vs. 30 cm *** p o0.001, 45 or 60 vs. 30 cm y p o0.001, 60 vs. 45 cm.
attributed to greater joint flexion (McNitt-Gray et al., 1993). The greater flexion naturally present in the hind limbs of rodents (Fig. 1) is also a likely explanation for the decreased forces in the hindlimbs of the rats when landing. Our force plate system was able to quantify peak vertical forces over the range of forces that occurred during the simulation of human impact exercise because of its ability to sample at 100 kHz, which commercially available small-force force plates systems cannot do. We tested several slower sampling frequencies and concluded that they cannot reliably capture the peak force at landing. Additionally, the ability of our force platform to capture vertical impact forces from more than one concurrent impact has several potential advantages over other methods. The platform can be placed on a bench top or floor without the need for the top of it to be level with the surrounding surface. Measuring the two forces concurrently reduces the variance inherent in measuring each force on separate drops, and this, in turn, reduces the number of landings required by the animals or subjects. The use of one platform instead of two parallel platforms reduces the cost
and space needed for 2 platforms. Two platforms would require more load cells to create a rigid planar surface; load cells contribute substantially to the cost of the system. Additionally, ensuring that both platforms are precisely the same height can be difficult and small differences in height could increase error in the results. Furthermore, although rats could be reliably dropped from low heights so that they land with prescribed feet on each platform, this is not the case from heights of 45 to 60 cm. Another benefit of one surface is that it allows direct measurement of the total vertical force from the impact of the entire rat and this measure adds to the accuracy of the results. Although the size of our platform was large enough to accommodate safe landings by young rats, it might be insufficient for bigger rats. A larger platform, with the load cells spread proportionately, could accommodate larger animals. Other small animal force platforms record three-dimensional forces, which can provide useful additional data for some research and cannot be measured by our system. However, when vertical forces are the sole dimension of interest, our platform is efficient and accurate. Errors in measurement included the estimation of the center of impact force on a foot from the center of an ink spot. We observed that relaxed rats landed with lowered heads and more weight on their forefeet. All the rats easily habituated to the dropping procedure but some found the ink pad to be stressful, which resulted in some variability in flight posture. Although the rats used were inbred, born on the same day and handled by the same
ARTICLE IN PRESS 2738
J.M. Welch et al. / Journal of Biomechanics 42 (2009) 2734–2738
person in as identical a manner as possible, variability in anxiety was observed and a likely reason for the high variability in landing stance (Table 3). The three-load cell force platform allowed quantification of the vertical ground reaction forces produced by the fore and hindfeet of rats, whether they were standing on, or dropped onto, the platform. This simple and inexpensive platform is ideal for accurately quantifying landing forces when rodents are used as models to test the effects of impact exercise on the musculoskeletal system.
Conflict of interest statement None of the authors of this manuscript has any conflict of interest pertaining to the results of this study. We do not have any financial and personal relationships with other people or organizations that could inappropriately influence the results in this paper.
Acknowledgement We would like to thank John Kozey for useful comments on a draft of this manuscript. References Akhter, M.P., Cullen, D.M., Pedersen, E.A., Kimmel, D.B., Recker, R.R., 1998. Bone response to in vivo mechanical loading in two breeds of mice. Calcif. Tissue Int. 63, 442–449. Bassey, E.J., Ramsdale, S.J., 1995. Weight-bearing exercise and ground reaction forces: a 12-month randomized controlled trial of effects on bone mineral density in healthy postmenopausal women. Bone 16, 469–476. Frost, H.M., 1987. Bone ‘‘mass’’ and the ‘‘mechanostat’’: a proposal. Anat. Rec. 219, 1–9. ¨ Heinonen, A., Kannus, P., Sievanen, H., Oja, P., Pasanen, M., Rinne, M., Uusi-Rasi, K., Vuori, I., 1996. Randomized controlled trial of effect of high-impact exercise on selected risk factors of osteoporotic fractures. Lancet 348, 1343–1347. ¨ ¨ ainen, H., Perttunen, J., Kannus, P., 2001. Mineral Heinonen, A., Sievanen, H., Kyrol¨ mass, size, and estimated mechanical strength of triple jumpers’ lower limb. Bone 29, 279–285. Honda, A., Nagasawa, S., Umemura, Y., 2003a. The effect of low-frequency jumptraining on bone mass and strength in the ovariectomized rats. Jpn. J. Phys. Fitness Sports Med. 52, 65–73. Honda, A., Sogo, N., Nagasawa, S., Shimizu, T., Umemura, Y., 2003b. High-impact exercise strengthens bone in osteopenic ovariectomized rats with the same outcome as Sham rats. J. Appl. Physiol. 95, 1032–1037. Honda, A., Umemura, Y., Nagasawa, S., 2001. Effect of high-impact and lowrepetition training on bones in ovariectomized rats. J. Bone Miner. Res. 16, 1688–1693.
Howard, C.S., Blakeney, D.C., Medige, J., Moy, O.J., Peimer, C.A., 2000. Functional assessment in the rat by ground reaction forces. J. Biomech. 33, 751–757. Judex, S., Zernicke, R.F., 2000. High-impact exercise and growing bone: relation between high strain rates and enhanced bone formation. J. Appl. Physiol. 88, 2183–2191. Kodama, Y., Umemura, Y., Nagasawa, S., Beamer, W.G., Donahue, L.R., Rosen, C.J., Baylink, D.J., Farley, J.R., 2000. Exercise and mechanical loading increase periosteal bone formation and whole bone strength in C57BL/6J mice but not in C3H/Hej mice. Calcif. Tissue Int. 66, 298–306. Lee, K.C., Maxwell, A., Lanyon, L.E., 2002. Validation of a technique for studying functional adaptation of the mouse ulna in response to mechanical loading. Bone 31, 407–412. McKay, H., Tsang, G., Heinonen, A., Mackelvie, K., Sanderson, D., Khan, K.M., 2005. Ground reaction forces associated with an effective elementary school based jumping intervention. Br. J. Sports Med. 39, 10–14. McNitt-Gray, J.L., Yokoi, T., Millward, C., 1993. Landing strategy adjustments made by female gymnasts in response to drop height and mat composition. J. Appl. Biomech. 9, 173–190. Nagasawa, S., Umemura, Y., 2002. Bone hypertrophy in rats: effects of jump number and height. Adv. Exercise Sports Physiol. 8, 87–92. Oggero, E., Pagnacco, G., Morr, D.R., Barnes, S.Z., Berme, N., 1997. The mechanics of drop landing on a flat surface—a preliminary study. Biomed. Sci. Instrum. 33, 53–58. Palombaro, K.M., 2005. Effects of walking-only interventions on bone mineral density at various skeletal sites: a meta-analysis. J. Geriatr. Phys. Ther. 28, 102–107. Quatman, C.E., Ford, K.R., Myer, G.D., Hewett, T.E., 2006. Maturation leads to gender differences in landing force and vertical jump performance: a longitudinal study. Am. J. Sports Med. 34, 806–813. Seegmiller, J.G., McCaw, S.T., 2003. Ground reaction forces among gymnasts and recreational athletes in drop landings. J. Athletic Training 38, 311–314. Turner, C.H., Roeder, R.K., Wieczorek, A., Foroud, T., Liu, G., Peacock, M., 2001. Variability in skeletal mass, structure, and biomechanical properties among inbred strains of rats. J. Bone Miner. Res. 16, 1532–1539. Umemura, Y., Baylink, D.J., Wergedal, J.E., Mohan, S., Srivastava, A.K., 2002a. A time course of bone response to jump exercise in C57BL/6J mice. J. Bone Miner. Metab. 20, 209–215. Umemura, Y., Ishiko, T., Tsujimoto, H., Miura, H., Mokushi, N., Suzuki, H., 1995. Effects of jump training on bone hypertrophy in young and old rats. Int. J. Sports Med. 16, 364–367. Umemura, Y., Ishiko, T., Yamauchi, T., Kurono, M., Mashiko, S., 1997. Five jumps per day increase bone mass and breaking force in rats. J. Bone Miner. Res. 12, 1480–1485. Umemura, Y., Sogo, N., Honda, A., 2002b. Effects of intervals between jumps or bouts on osteogenic response to loading. J. Appl. Physiol. 93, 1345–1348. Webb, A.A., Gowribai, K., Muir, G.D., 2003. Fischer (F-344) rats have different morphology, sensorimotor and locomotor abilities compared to Lewis, LongEvans, Sprague–Dawley and Wistar rats. Behav. Brain Res. 144, 143–156. Welch, J.M., Turner, C.H., Devareddy, L., Arjmandi, B.H., Weaver, C.M., 2008. High impact exercise is more beneficial than dietary calcium for building bone strength in the growing rat skeleton. Bone 42, 660–668. Welch, J.M., Weaver, C.M., Turner, C.H., 2004. Adaptations to freefall impact are different in the shafts and bone ends of rat forelimbs. J. Appl. Physiol. 97, 1859–1865. Zumwalt, A.C., Hamrick, M., Schmitt, D., 2006. Force plate for measuring the ground reaction forces in small animal locomotion. J. Biomech. 39, 2877–2881.
Contents lists available at ScienceDirect
Journal of Biomechanics journal homepage: www.elsevier.com/locate/jbiomech www.JBiomech.com
Force platform for rats measures fore and hind forces concurrently Jo M. Welch a,b,, Jeremie A. Wade c, Ben M. Hillberry c, Connie M. Weaver a a b c
Foods and Nutrition, Purdue University, West Lafayette, IN, USA Health and Human Performance, Dalhousie University, Halifax, NS, Canada B3H 3J5 Mechanical Engineering, Purdue University, West Lafayette, IN, USA
a r t i c l e in f o
a b s t r a c t
Article history: Accepted 11 August 2009
Animal models are commonly used to test the efficacy of impact loading regimens on bone strength. We designed an inexpensive force platform to concurrently measure the separate peak vertical impact forces produced by the fore and hindfeet of immature F-344 rats when dropped onto the platform. The force platform consisted of three load cells placed in a triangular pattern under a flat plate. Rats were dropped from heights of 30, 45 and 60 cm onto the platform so that they landed on all four feet concurrently. The peak vertical impact forces produced by the feet of the rats were measured using a sampling frequency of 100 kHz. The location of each foot at landing relative to the load cells, and the force received by each load cell were combined in a series of static equations to solve for the vertical impact forces produced by the fore and hindfeet. The forces produced by feet when rats stood on the single platform were similarly determined. The forces exerted separately by the fore and hindfeet of young rats when landing on the plate as a ratio to standing forces were then calculated. Rats when standing bore more weight on their hindfeet but landed with more weight on their forefeet, which provides rationale for the greater response to landing forces of bones in the forelimbs than those in the hindlimbs. This system provided a useful method to simultaneously measure peak vertical impact forces in fore and hindfeet in rats. & 2009 Elsevier Ltd. All rights reserved.
Keywords: Force platform Ground reaction forces Rat Bone Impact exercise
1. Introduction The minimum ground reaction force (GRF) that is required to activate beneficial bone remodelling appears to differ with age, propensity for bone strength, and type of habitual exercise. Knowledge of which exercise protocols might trigger bone change during growth or aging, and under various other conditions could reduce rates of fracture in humans. Bone responds positively to forces that meet or exceed a threshold (Frost, 1987). High impact activities such as those delivered by the track and field sport of triple jump are highly osteogenic (Heinonen et al., 2001), while low impact activities, such as walking, produce little change in bone (Palombaro, 2005). Axial impact forces can improve long bone strength (Welch et al., 2004; Heinonen et al., 1996) and are typically measured using a force platform. Resultant force values are typically normalized using standing forces and reported as either a multiple of body weight (Quatman et al., 2006) or as the ratio of forces, N landing/N standing (Bassey and Ramsdale, 1995). The use of rodents as models to test exercise protocols for osteogenicity offers the advantages of precise control of dietary, Corresponding author at: School of Health and Human Performance, 6230 South Street, Dalhousie University, Halifax, Nova Scotia, Canada B3H 3J5. Tel.: + 1 902 494 2475; fax: + 1 902 494 5120. E-mail address: [email protected] (J.M. Welch).
0021-9290/$ - see front matter & 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.jbiomech.2009.08.002
exercise, and environmental conditions, the ability to measure bone strength directly, and large reductions in the experimental time needed to detect changes in bone. In animals, the effects on bone of impact from jumping, or from freefalling onto a surface, has been reported in rats (Umemura et al., 1997, 1995, 2002b; Honda et al., 2001, 2003b, 2003a; Nagasawa and Umemura, 2002; Welch et al., 2004), mice (Lee et al., 2002; Kodama et al., 2000; Umemura et al., 2002a), and poultry (Judex and Zernicke, 2000). Methods used to measure GRF in humans require modification for use in quadrupeds. Several methods to quantify GRFs during rodent locomotion have been reported. Howard and coworkers (Howard et al., 2000) walked adult rats in two directions through a narrow channel that allowed only ipsilateral limbs to pass over the force platform. Webb et al. (2003); Howard et al. (2000) trotted young rats over a raised runway with three small platforms embedded beneath it to independently capture the forces from each limb. Zumwalt et al. (2006) walked mice over a force plate that was modified so that only a small area was sensitive to applied forces. Commercially available small-force force plates systems, such as the HE6 6, (Advanced Mechanical Technology, Incorporated, Watertown, MA) that was used by Zumwalt et al. (2006) have maximum sampling frequencies that are too low to capture the peak vertical forces present at an impact from drop jumps and therefore cannot accurately capture
ARTICLE IN PRESS J.M. Welch et al. / Journal of Biomechanics 42 (2009) 2734–2738
the peak landing forces. Quantifying the ground reaction force received by one leg when landing from a height requires a different method. Methods commonly used for measuring these forces in humans include two parallel force platforms (Oggero et al., 1997; Quatman et al., 2006), or landing with one foot on the platform and the other on the floor beside the platform (Seegmiller and McCaw, 2003). Although humans bear similar weight on both feet when landing from drop jumps (Quatman et al., 2006), quadrupeds display different proportions of weight on their fore and hindfeet (Welch et al., 2004). Fore and hind differences are also specific to each strain of rat, and to the activity that they are performing (Webb et al., 2003). Rodents also differ from humans in that they cannot be expected to land oriented as the researcher prefers and most will only momentarily stand in the location where they landed. It is therefore important to be able to differentiate the peak force from subsequent ones and to differentiate between where the feet landed and subsequent steps. A successful force platform to assess vertical impact in rats therefore requires a high sampling frequency and an adjustable landing area. In earlier experiments we examined the effects of freefall landings on the geometry and tissue properties of leg bones in growing rats that were dropped from 30, 60 cm (Welch et al., 2004) and 45 cm (Welch et al., 2008). Results indicated substantial osteogenic benefits from this activity in the forelimb bones but muted or absent responses in the hindlimbs. We therefore hypothesized that the hindlimbs received lower impact forces than did the forelimbs. To test this hypothesis, we constructed a force platform that allowed concurrent measurement of the fore and hind peak vertical impact forces in rats at landing from a fixed drop height. Preliminary GRF data from this method were previously reported for the 30-, 60-cm (Welch et al., 2004) and 45-cm (Welch et al., 2008) drop distances. Presently, we are reporting the description of the force platform system used to determine the vertical ground reaction forces exerted on the fore and hindfeet of rats landing from heights of 30, 45, and 60 cm.
2735
platform when the rats landed from the different heights. Additionally, these standard weights were placed at other locations on the force plate and the values at the load cells compared favourably with calculated values. The calibration curves included zero and therefore the system was tested over the full range of values obtained during the experiment. When the rats were dropped, a solid boundary was placed around and nearly touching the force plate (not shown) as we observed that it improved the relaxation of the animals and facilitated their retention on the lab bench. Standard graph paper was attached to the top of the force platform so that the locations of the feet after the drop could be marked and these locations could then be recorded as coordinates relative to the load cell locations. The rats can be dropped in any orientation so long as the locations of the paw landings relative to the load cells are included in the equations.
2.3. Protocol Fischer-344 rats tend to be a nervous strain and therefore the rats were habituated to human contact and procedures over a three week period. Each rat was picked up daily by a handler who grasped its body with a bare hand, lifted, petted, and carried it about the room. After one week, rats began habituation to the dropping protocol. An initial three very low drops progressed in number, height, and increased frequency until the final regimen assigned to each rat was reached. Rats were randomized, based on body weight, to 30, 45, or 60 cm dropping heights (n= 4 each group). All force platform measurements for all rats were recorded within a 24-h period, with different rats selected at each drop height to avoid either changes in body size that would occur due to growth if the experiment took multiple days, or the physical stress from 30 instead of 10 landings per rat on one day. We previously observed that most trained rats remained relaxed, as demonstrated in inverted U-shaped position of the rat in Fig. 2, for at least 15 freefalls per session. On the data collection day, each rat was placed on an ink pad containing washable ink and then lifted until its feet reached the prescribed height. It was dropped so that it landed on all feet approximately concurrently. Non-simultaneous landings were audible on the aluminum platform; results from those drops were excluded and additional drops performed. The coordinates for the center of each footprint were measured from the ink prints on the graph paper (Fig. 3). The peak values of the forces recorded concurrently by each load cell at the
2. Materials and methods 2.1. Animals Twelve female Fischer-344 rats, aged 46 days, were purchased from Harlan Sprague–Dawley, Inc. (Indianapolis, IN, USA). This strain of inbred rats was the same strain, sex, and age as those used previously to test the efficacy of the freefall impact protocol on osteogenicity (Welch et al., 2004, 2008). This strain is more fragile-boned than others (Turner et al., 2001) and therefore is ideal for testing the effects of physical forces on bone because fragile-boned rodents are more responsive to such forces (Akhter et al., 1998). On the day the measurements were made, the rats were 67 days old, had a mean mass of 140 g, and were growing at a mean rate of 1.6 g d 1. Animal protocols and care were approved by the Purdue Animal Care and Use Committee of Purdue University.
Fig. 1. The three-load cell force platform: (A) force platform, (B) base plate, (C) load cell.
2.2. Apparatus A 17.78 cm 17.78 cm aluminium force platform was attached above three 2.27 kg load cells (P/N AL311AT; Honeywell Sensotec, Columbus, OH, USA). An 11.43 cm 11.43 cm steel base plate (2.0 kg) was attached to the bottom of the load cells. The force plate system is shown in Fig. 1. When assembling the system, the locations of the load cells form a triangle but their precise locations can vary, as can the relative locations of the load cells to each other. The load cells were connected to three analog channels via a 68-pin shielded I/O connector block (P/N SCB68, National Instruments, Austin, TX, USA). A shielded cable connected this breakout box to a multifunction DAQ card (P/N PCI-6034E, National Instruments). Data were sampled at 100 kHz by a custom program using Labview 7 Express (National Instruments). Before data collection, to test for linearity and accuracy, calibration of each load cell was performed while connected to the platform. Five standard weights of 1, 2, 3, 4 and 5 lb, corresponding to forces of approximately 4, 9, 13, 18, 22 N, were placed sequentially directly over each load cell and calibration curves were constructed from this data for each load cell. These calibration weights were selected because they spanned the range of forces received by the
Fig. 2. A rat in freefall flight. A relaxed rat travels with its legs and back forming an inverted U-shape, and appears to look for its landing surface.
ARTICLE IN PRESS 2736
J.M. Welch et al. / Journal of Biomechanics 42 (2009) 2734–2738 averaged and assumed to be the same for all rats. Standing forefeet weight for each rat was then calculated as follows: Standing forefeet weight ¼ rat mass g k where g = 9.81 m s 1, and k is the proportion of rat mass on the forefeet when standing. The peak vertical impact forces produced by fore and hindfeet when a rat was dropped onto the force platform were standardized by dividing each rat’s fore and hindfoot forces by body weight, respectively, when standing. Contralateral legs were assumed to bear equal force. 2.4. Statistical analysis Differences between groups were tested by ANOVA using SPSS version 14.0. Pairwise comparisons between means were made post hoc using the Tukey’s HSD test. An a level was set at po 0.05 for all tests. We hypothesized that rats could modify their impacts by landing from different heights with different foot stances. To address this, the distances between fore, hind, and fore to hindfeet at the moment of landing from the three drop heights were tested for difference using ANOVA with body mass as a covariate. Coefficients of variation (CV) for the peak landing forces and landing stances were calculated from the pooled variance of four individual variances per treatment; variances for individual rats were computed from the 10 replications per rat.
3. Results Fig. 3. Rat footprints on the landing grid. X and Y coordinates for each foot were determined from the corner of the platform nearest the left anterior of the rat. All rats were dropped to face the same direction.
Table 1 Nomenclature and equations used in calculation of fore and hindfeet forces. Nomenclature C F x y
Load cell peak force Foot peak vertical impact force Horizontal coordinate Vertical coordinate
Subscripts a–c 1 2 3 4
Load cells a, b, c Right forefoot Right hindfoot Left hindfoot Left forefoot
Equations (1)
xfore = 12(x1 + x4)
(2)
yfore = 12(y1 + y4)
(3)
xhind = 12(x2 +x3)
(4)
yhind = 12(y2 + y3) Ffore(xfore xa)+ F2(x2 xa)+ F3(x3 xa) = Cb(xb xa) +Cc(xc xa) Ffore(yfore ya) + F2(y2 ya) + F3(y3 ya) = Cb(yb ya) +Cc(yc ya) Ffore + F2 +F3 = Ca +Cb + Cc
(5) (6) (7)
moment of impact were entered into the custom program. Feet coordinates were combined with load cell maximal forces to calculate the peak vertical impact force using the following method. (Nomenclature and equations are defined in Table 1.) Spatial locations for the midpoint of the fore and hindfeet were calculated and denoted as xfore, yfore, xhind, and yhind (Eqs. (1–4)). F1, F2, F3, and F4 represented the vertical impact forces on each foot at landing. F1 and F4 represented the forefeet and F2 and F3 represented the hindfeet. To reduce four forces to three force unknowns to be used with three static equations, a force for the forefeet, Ffore, occurring at (xfore, yfore), was created. Three simultaneous equations (Eqs. (5–7)) were solved to obtain values for Ffore, F2, and F3, where the latter two variables are the GRFs of each hindfoot. The corresponding force for the hindfeet, Fhind, occurring at (xhind, yhind), was calculated by subtracting Ffore from the total impact forces. The proportion of mass borne by the fore and hindfeet of the rats when standing was measured by placing rats from each group on the force platform. Mean fore and hind values from a total of 10 repetitions using six rats were used to calculate the proportion of weight borne by the fore and hindfeet. Other repetitions were discarded because the animals would not stand still. The proportions of forces borne by the fore and hindfeet from this subset of rats were
The specially designed force plate with associated methodology and customized analysis was used to quantify the landing forces of rats dropped from 30, 45, and 60 cm. The forefeet of rats were measured as one force, and both hindfeet together as another, separate force. Standing rats bore an average of 41.6% of their weight on their forefeet. Rats dropped from 30, 45, and 60 cm bore 65.8%, 59.3%, and 60.7% of the impact forces on their forefeet, respectively. Peak vertical impact forces from 10 drops per rat were averaged and the mean forces for each rat are presented in Table 2. At landing, the forces borne by the feet relative to those borne by the same feet when standing were significantly greater in the fore than the hindfeet (Table 2, Fig. 4). In the forefeet each increment of drop height resulted in significantly greater landing forces. However, in the hindfeet, the forces borne at 45 and 60 cm were greater than at 30 cm, but were not significantly different from each other. Total landing force had a much lower variability between drops and between rats in the same groups than did fore and hind forces. Stance at landing did not differ between groups. The CV for the peak vertical landing force of the fore, hind, and total feet, and the landing stances, are presented in Table 3.
4. Discussion A method to measure peak vertical impact forces in the fore and hindfeet of rats dropped onto one force platform is described. This simple bench top system is useful in quantifying vertical ground reaction forces in rodents and is especially well suited to measure these forces in animals used as models for impact exercise research. We were able to determine that rats, when landing on four feet concurrently from a height, bore more weight on their forefeet than hindfeet. This knowledge was useful to explain the previous findings that increased strength in the forelimb bones in young rats that were dropped 10 times per day from 30 or 60 cm for eight weeks correlated with the greater forces borne by the front limbs (Welch et al., 2004). Peak impact forces increased linearly with drop height in the forelimbs but in the hindlimbs, forces approached an asymptote at the 45 cm height. Drop jumps by gymnasts have elicited a linear increase in vertical GRF with height of drop (McKay et al., 2005; Seegmiller and McCaw, 2003) through a range of low heights but this linearity was blunted at greater jump heights, which was
ARTICLE IN PRESS J.M. Welch et al. / Journal of Biomechanics 42 (2009) 2734–2738
2737
Table 2 Forces measured and ground reaction forces calculated from rat freefalls onto force platforma. Distance dropped
Landing F, all feet (N) Landing F, all feet/rat mass (N/g) Fraction landing F, front feet (%) Landing F, each front foot (N) Landing F, front foot/rat mass (N/g) Landing F, each hindfoot (N) Landing F, back foot/rat mass (N/g) Mass of rat (g) Fraction of standing F, calculated, front feet (%)b Front foot forces, landing/standingc Hindfoot forces, landing/standingd
30 cm
45 cm
60 cm
10.15 70.08 0.07570.001 65.80 71.13 3.34 70.08 0.025 70.000 1.74 70.05 0.013 70.000 136 73 41.56 12.04 70.24 4.46 70.20
14.15 70.55*** 0.099 70.001*** 59.29 70.62* 4.19 70.18* 0.029 70.001*** 2.88 70.10*** 0.020 70.001*** 143 74 41.56 14.43 70.32** 7.11 70.13***
15.917 0.35***y 0.1127 0.002***yyy 60.747 1.99 4.84 7 0.25*** 0.0347 0.001***yyy 3.117 0.11*** 0.0227 0.001*** 1427 3 41.56 16.747 0.52***yy 7.77 0.43***
Different from 30 cm: *p o 0.05, **p o0.01, and ***p o0.001. 60 cm is different from 45 cm: ypo 0.05, yyp o0.01, and yyyp o 0.001. a
Each data point represents the mean values from four rats7SE. Constant calculated from 10 repetitions with six rats and extrapolated to all rats. Standing weight or force on forefeet calculated from rat mass g k, where g= 9.81 m s 1, and k is the proportion of rat mass on the forefeet. d Standing weight or force on hindfeet calculated from rat mass g (1 k), where g = 9.81 m s 1, and k is the proportion of rat mass on the forefeet. b c
Peak impact force (N*N-1)
19
*** †
16
Table 3 Coefficients of variation (%) for the rat force platform.
**
Drop distance (cm) 30 (n= 4)
45 (n= 4)
60 (n= 4)
Peak vertical forces Total (four feet) Forefeet Hind feet
5.2 10.9 20.1
6.9 11.4 18.4
10.2 14.7 19.3
Landing stancea Fore left to right Hind left to right Mean fore to hind
11.9 20.0 14.3
14.6 13.2 9.7
17.5 17.2 8.9
13 10
***
7
***
4 1
a
0
30 45 Drop distance (cm)
Landing stance refers to the distances between feet at landing.
60
Fig. 4. Peak vertical impact forces produced by rats dropped from heights of 30, 45, and 60 cm. Data are means 7 SD; n= 4. Solid line represents forefoot impact force per forefoot weight; dashed line represents hindfoot impact force per hindfoot weight. For each line, statistical differences are denoted as follows: ** p o 0.01, vs. 30 cm *** p o0.001, 45 or 60 vs. 30 cm y p o0.001, 60 vs. 45 cm.
attributed to greater joint flexion (McNitt-Gray et al., 1993). The greater flexion naturally present in the hind limbs of rodents (Fig. 1) is also a likely explanation for the decreased forces in the hindlimbs of the rats when landing. Our force plate system was able to quantify peak vertical forces over the range of forces that occurred during the simulation of human impact exercise because of its ability to sample at 100 kHz, which commercially available small-force force plates systems cannot do. We tested several slower sampling frequencies and concluded that they cannot reliably capture the peak force at landing. Additionally, the ability of our force platform to capture vertical impact forces from more than one concurrent impact has several potential advantages over other methods. The platform can be placed on a bench top or floor without the need for the top of it to be level with the surrounding surface. Measuring the two forces concurrently reduces the variance inherent in measuring each force on separate drops, and this, in turn, reduces the number of landings required by the animals or subjects. The use of one platform instead of two parallel platforms reduces the cost
and space needed for 2 platforms. Two platforms would require more load cells to create a rigid planar surface; load cells contribute substantially to the cost of the system. Additionally, ensuring that both platforms are precisely the same height can be difficult and small differences in height could increase error in the results. Furthermore, although rats could be reliably dropped from low heights so that they land with prescribed feet on each platform, this is not the case from heights of 45 to 60 cm. Another benefit of one surface is that it allows direct measurement of the total vertical force from the impact of the entire rat and this measure adds to the accuracy of the results. Although the size of our platform was large enough to accommodate safe landings by young rats, it might be insufficient for bigger rats. A larger platform, with the load cells spread proportionately, could accommodate larger animals. Other small animal force platforms record three-dimensional forces, which can provide useful additional data for some research and cannot be measured by our system. However, when vertical forces are the sole dimension of interest, our platform is efficient and accurate. Errors in measurement included the estimation of the center of impact force on a foot from the center of an ink spot. We observed that relaxed rats landed with lowered heads and more weight on their forefeet. All the rats easily habituated to the dropping procedure but some found the ink pad to be stressful, which resulted in some variability in flight posture. Although the rats used were inbred, born on the same day and handled by the same
ARTICLE IN PRESS 2738
J.M. Welch et al. / Journal of Biomechanics 42 (2009) 2734–2738
person in as identical a manner as possible, variability in anxiety was observed and a likely reason for the high variability in landing stance (Table 3). The three-load cell force platform allowed quantification of the vertical ground reaction forces produced by the fore and hindfeet of rats, whether they were standing on, or dropped onto, the platform. This simple and inexpensive platform is ideal for accurately quantifying landing forces when rodents are used as models to test the effects of impact exercise on the musculoskeletal system.
Conflict of interest statement None of the authors of this manuscript has any conflict of interest pertaining to the results of this study. We do not have any financial and personal relationships with other people or organizations that could inappropriately influence the results in this paper.
Acknowledgement We would like to thank John Kozey for useful comments on a draft of this manuscript. References Akhter, M.P., Cullen, D.M., Pedersen, E.A., Kimmel, D.B., Recker, R.R., 1998. Bone response to in vivo mechanical loading in two breeds of mice. Calcif. Tissue Int. 63, 442–449. Bassey, E.J., Ramsdale, S.J., 1995. Weight-bearing exercise and ground reaction forces: a 12-month randomized controlled trial of effects on bone mineral density in healthy postmenopausal women. Bone 16, 469–476. Frost, H.M., 1987. Bone ‘‘mass’’ and the ‘‘mechanostat’’: a proposal. Anat. Rec. 219, 1–9. ¨ Heinonen, A., Kannus, P., Sievanen, H., Oja, P., Pasanen, M., Rinne, M., Uusi-Rasi, K., Vuori, I., 1996. Randomized controlled trial of effect of high-impact exercise on selected risk factors of osteoporotic fractures. Lancet 348, 1343–1347. ¨ ¨ ainen, H., Perttunen, J., Kannus, P., 2001. Mineral Heinonen, A., Sievanen, H., Kyrol¨ mass, size, and estimated mechanical strength of triple jumpers’ lower limb. Bone 29, 279–285. Honda, A., Nagasawa, S., Umemura, Y., 2003a. The effect of low-frequency jumptraining on bone mass and strength in the ovariectomized rats. Jpn. J. Phys. Fitness Sports Med. 52, 65–73. Honda, A., Sogo, N., Nagasawa, S., Shimizu, T., Umemura, Y., 2003b. High-impact exercise strengthens bone in osteopenic ovariectomized rats with the same outcome as Sham rats. J. Appl. Physiol. 95, 1032–1037. Honda, A., Umemura, Y., Nagasawa, S., 2001. Effect of high-impact and lowrepetition training on bones in ovariectomized rats. J. Bone Miner. Res. 16, 1688–1693.
Howard, C.S., Blakeney, D.C., Medige, J., Moy, O.J., Peimer, C.A., 2000. Functional assessment in the rat by ground reaction forces. J. Biomech. 33, 751–757. Judex, S., Zernicke, R.F., 2000. High-impact exercise and growing bone: relation between high strain rates and enhanced bone formation. J. Appl. Physiol. 88, 2183–2191. Kodama, Y., Umemura, Y., Nagasawa, S., Beamer, W.G., Donahue, L.R., Rosen, C.J., Baylink, D.J., Farley, J.R., 2000. Exercise and mechanical loading increase periosteal bone formation and whole bone strength in C57BL/6J mice but not in C3H/Hej mice. Calcif. Tissue Int. 66, 298–306. Lee, K.C., Maxwell, A., Lanyon, L.E., 2002. Validation of a technique for studying functional adaptation of the mouse ulna in response to mechanical loading. Bone 31, 407–412. McKay, H., Tsang, G., Heinonen, A., Mackelvie, K., Sanderson, D., Khan, K.M., 2005. Ground reaction forces associated with an effective elementary school based jumping intervention. Br. J. Sports Med. 39, 10–14. McNitt-Gray, J.L., Yokoi, T., Millward, C., 1993. Landing strategy adjustments made by female gymnasts in response to drop height and mat composition. J. Appl. Biomech. 9, 173–190. Nagasawa, S., Umemura, Y., 2002. Bone hypertrophy in rats: effects of jump number and height. Adv. Exercise Sports Physiol. 8, 87–92. Oggero, E., Pagnacco, G., Morr, D.R., Barnes, S.Z., Berme, N., 1997. The mechanics of drop landing on a flat surface—a preliminary study. Biomed. Sci. Instrum. 33, 53–58. Palombaro, K.M., 2005. Effects of walking-only interventions on bone mineral density at various skeletal sites: a meta-analysis. J. Geriatr. Phys. Ther. 28, 102–107. Quatman, C.E., Ford, K.R., Myer, G.D., Hewett, T.E., 2006. Maturation leads to gender differences in landing force and vertical jump performance: a longitudinal study. Am. J. Sports Med. 34, 806–813. Seegmiller, J.G., McCaw, S.T., 2003. Ground reaction forces among gymnasts and recreational athletes in drop landings. J. Athletic Training 38, 311–314. Turner, C.H., Roeder, R.K., Wieczorek, A., Foroud, T., Liu, G., Peacock, M., 2001. Variability in skeletal mass, structure, and biomechanical properties among inbred strains of rats. J. Bone Miner. Res. 16, 1532–1539. Umemura, Y., Baylink, D.J., Wergedal, J.E., Mohan, S., Srivastava, A.K., 2002a. A time course of bone response to jump exercise in C57BL/6J mice. J. Bone Miner. Metab. 20, 209–215. Umemura, Y., Ishiko, T., Tsujimoto, H., Miura, H., Mokushi, N., Suzuki, H., 1995. Effects of jump training on bone hypertrophy in young and old rats. Int. J. Sports Med. 16, 364–367. Umemura, Y., Ishiko, T., Yamauchi, T., Kurono, M., Mashiko, S., 1997. Five jumps per day increase bone mass and breaking force in rats. J. Bone Miner. Res. 12, 1480–1485. Umemura, Y., Sogo, N., Honda, A., 2002b. Effects of intervals between jumps or bouts on osteogenic response to loading. J. Appl. Physiol. 93, 1345–1348. Webb, A.A., Gowribai, K., Muir, G.D., 2003. Fischer (F-344) rats have different morphology, sensorimotor and locomotor abilities compared to Lewis, LongEvans, Sprague–Dawley and Wistar rats. Behav. Brain Res. 144, 143–156. Welch, J.M., Turner, C.H., Devareddy, L., Arjmandi, B.H., Weaver, C.M., 2008. High impact exercise is more beneficial than dietary calcium for building bone strength in the growing rat skeleton. Bone 42, 660–668. Welch, J.M., Weaver, C.M., Turner, C.H., 2004. Adaptations to freefall impact are different in the shafts and bone ends of rat forelimbs. J. Appl. Physiol. 97, 1859–1865. Zumwalt, A.C., Hamrick, M., Schmitt, D., 2006. Force plate for measuring the ground reaction forces in small animal locomotion. J. Biomech. 39, 2877–2881.