バイオメカニズム学会誌，Vol. 38，No.2（2014）

解 説

Running-specific prostheses: The history, mechanics, and controversy

Hiroaki HOBARA1† 1Digital Human Research Center, National Institute of Advanced Industrial Science and (AIST)

Abstract : Recent developments in carbon fiber running-specific prostheses (RSPs) have allowed individuals with lower extremity (ILEA) to regain the functional capability of running. There are many amputee sprinters who are now able to run faster and achieve longer jumps than able-bodied athletes. However, ironically, this phenomenon has raised a debate in the scientific community regarding the potential advantages or disadvantages of RSPs in athletic ILEA compared to able-bodied counterparts in running. This article describes the history, classification, and regulations of RSPs, and current world records in athletic ILEA. Finally, a debate regarding the advantages or disadvantages of RSPs is presented.

Key Words : Prostheses, Amputees, Sprinting, ,

1. INTRODUCTION running biomechanics in ILEA and the biomechanical function Recent technical developments in carbon fiber running-specific of prostheses during this activity. However, due to the lack of prostheses (RSPs) with energy storing capabilities have allowed studies on running in ILEA and the dearth of information on individuals with lower extremity amputation (ILEA) to compete RSPs, quantification of biomechanical parameters in ILEA during at levels achieved never before. Additionally, RSPs have attracted running using RSPs is scarce. more and more ILEA to running as a form of exercise and athletic This article describes the history, classification, and regulations competition. In Sainsbury’s Anniversary Games at the Olympic of RSPs, and current world records of athletic ILEA. After Stadium on July 28, 2013, Alan Fonteles Oliveira (Brazil) and discussing the advantages and disadvantages of RSPs, several Richard Browne (USA), who are the best amputee sprinters in the suggestions for future studies are presented. It is my hope that world, set new world-best times in the 100-m T43 and T44 classes, this paper will shed light on the role of RSPs in the area of respectively. Oliveira broke his own T43 record, achieving victory biomechanics. This paper presents some of the findings and ideas with a time of 10.57 [s], while Browne set a new T44 world record published in a previous publication1). of 10.75 [s] and finished second. Furthermore, the participation of a South African double-amputee sprinter () in an 2. HISTORY OF RUNNING-SPECIFIC PROSTHESES event with able-bodied sprinters (Men’s 400[m] race) at the 2012 After the invention of the Solid and Cushioned London Olympics is still fresh in our minds. (SACH) (Ohio Willow Wood, Ohio, USA) in the late Not surprisingly, there are a lot of amputee sprinters who are 1950s, prosthetic foot designs and materials changed little for able to run faster and achieve longer jumps than able-bodied approximately 20-30 years2). According to the previous studies3), athletes. The phenomenon may exemplify how amputee sprinters the usefulness of lower- prostheses improved tremendously work hard with high motivation, and how current prostheses in the 1980s, when advances in composite materials flooded the have advanced. Therefore, developments of improved training/ prosthetics industry. Carbon composite materials, used extensively rehabilitation regimes and prosthetic designs to promote running in the aerospace industry, brought lightness, durability, and within this population requires a detailed understanding of strength to the design of prosthetic feet, pylons, and sockets2-4). In 1984, Van Phillips, an American inventor of prostheses, created the “Flex-Foot®” (Figure 1) made of carbon graphite. The innovative 2013 年 12 月 24 日受付 † artificial foot allowed users to store and then return elastic energy Waterfront 3F, DHRC, 2-3-26, Aomi, Koto-ku, Tokyo, 135-0064, JAPAN Hiroaki HOBARA during the ground-contact phase of gait. E-mail: [email protected] The Flex-Foot was first seen in elite sports at the 1988 Phone: +81-3-3599-8201 Fax: +81-3-5530-2066

ー 105 ー バイオメカニズム学会誌，Vol. 38，No.2（2014）

in 1989, his personal record in the 100[m] race was only 14.38 [s]. Later, when provided with the Flex-Foot prostheses, he won the gold medal at the Atlanta Paralympic Games in the men’s 100[m] race by setting a world record with a time of 11.36 [s]. Furthermore, he also won the gold medal in the 200[m] race with a time of 23.28 [s]. For the last 15 years, technical advances in prostheses have been a main factor in the increased performance of athletes with lower-limb . The use of materials such as carbon fiber, , and graphite has provided added strength and energy- Figure 1 Schematic representation of “Flex-Foot®” and prosthetic storage capabilities to prostheses while decreasing the weight of components (socket and liner) with representation of a residual prosthetic components6). Today, carbon fiber prostheses are most limb. The schema is based on transtibial (below-) amputees. popular in elite running and jumping events. These prostheses allow lower-limb amputees to actively participate in sporting activities including competitive sports4).

3. CLASSIFICATION AND REGULATIONS IN ATHLETICS According to the International Paralympic Committee (IPC) regulations, transtibial amputee runners were classified into two classes: T43 class (double below-knee amputees and other Figure 2 Typical examples of running-specific prostheses (A to F were athletes with impairments that are comparable to a double-below adapted from from Lechlar and Lilja, [29]). A: Flex-Foot® (Modular III; Össur), B: Flex-Sprint II (Össur), C: Flex-Sprint I knee amputation) and T44 class (any athlete with a lower limb (Össur), D: Flex-Sprint III® (Cheetah; Össur), E: Flex-RunTM impairment/s that meets minimum criteria for lower limb (Össur), F: Symes-Sprint (Össur), G: Cheetah Xtreme® (Össur, deficiency, impaired lower limb passive range of motion, impaired https://www.ossur.com), H: Cheetah Xtend® (Össur, https://www. lower limb muscle power, or length difference). Although ossur.com), I: 1E90 (Sprinter, OttoBock, http://www.ottobockus. com), J: 1C2 (C-Sprint®, http://www.ottobockus.com), K: Nitro official records in the two classes were separately recognized, (Freedom Innovation, http://www.freedom-innovations.com), T43 and 44 classes were integrated into the same race in 2012. L: CatapultTM (Freedom Innovation, http://www.freedom- However, in the 6th IPC Athletics World Championships, in Lyon, innovations.com), M: SP1100 (KATANA, IMASEN Engineering Corporation, http://www.imasengiken.co.jp/), N: Rabbit France, in 2013, the two classes competed separately in a race. (IMASEN Engineering Corporation). Note that transfemoral amputees were classified in the T42 class (single above-knee amputees and athletes with other impairments that are comparable to a single above-knee amputation) regardless Paralympic Games5). Four years later, the prosthetic heel was of unilateral or bilateral amputations. absent in some athletes5) creating the first sprint prosthesis2). In According to Rule 3.3.2 of the IPC official guidelines, a fact, the first specialized running foot, the Flex-Sprint I (Figure competitor’s artificial limb “must not create the unrealistic 2-C; Össur, Reykjavik, Iceland), was developed by eliminating the enhancement of stride length.” Furthermore, in the IPC regulations, heel portion and altering the stiffness configuration with the lay- there are also statements such as: “equipment and/or prosthetic up sequence of carbon while still maintaining the J-shaped outline components are commercially available to all athletes (i.e., of the carbon forefoot. There are now several different sprint foot prototypes that are purposely built by manufacturers exclusively designs available, all with a similar basic shape (Figure 2-A to N), for the use of a specific athlete should not be permitted).” which has changed little since 19922). Similarly, “equipment used contains materials or devices that store, The advent of carbon-fiber prostheses and RSPs shortened generate, or deliver energy and/or are designed to provide function approximately 1.5 [s] off the world record of 100[m] races in the to enhance performance beyond the natural physical capacity of T44 class (transtibial amputees) within 10 years (from 1988 to the athlete” shall not be permitted. Furthermore, athletes using a 1998). One of the best examples of this context is the Paralympic cannot use different lengths of prosthesis for different Games in Atlanta, Georgia, in 1996. Tony Volpentest, an American disciplines at the same competition (e.g. for the 100[m] and Paralympian athlete, was born with short malformed and the 400[m] race). However, it is possible for athletic ILEA to . When he initiated running using the walking prostheses

ー 106 ー バイオメカニズム学会誌，Vol. 38，No.2（2014）

use different prostheses for track events and field events if the Table 1 Current world records in the 100-, 200-, and 400-meter sprints prosthetic length is the same. In addition, the IPC has strict rules in able-bodied athletes (AB) and T43 and T44 classes (as of December 24, 2013). Data were adapted from the International for length of RSPs in runners with lower extremity amputation. Association of Athletics Federations (IAAF) website and the International Paralympic Committee (IPC) website. 4. WORLD RECORDS IN ATHLETICS The recent carbon-fiber prostheses along with hard training now enable athletes with amputations to run 100 m in just under 11[s]. The current world records in the 100, 200, and 400[m] sprints in T43 and T44 classes compared with the able-bodied world records, according to the IPC, are summarized in Table 1. Interestingly, many world records in track and field competitions were established in the last three years (2011 to 2013). In both, men and women, the T43 class has faster world records than the T44 class for 100, 200 and 400[m] sprints (Table 1). Furthermore, the women’s 100[m] T43 world record has come down by nearly five seconds in 14 years, and that of the T44 by just under two seconds. These results indicate that double-leg amputee sprinters may obtain more advantage than single-leg amputee sprinters from carbon-fiber prosthetics.

5. ADVANTAGE OR DISADVANTAGE? As described in the Introduction, a South African double- amputee sprinter took part in the 2011 World Championships and 2012 Olympics in London. In July 2011, he ran the 400[m] race in 45.07[s] in Lignano, Italy, passing the qualifying standard that allowed him to represent South Africa alongside the best able- bodied athletes at the 2011 World Championships in August 2011. Regardless of what’s going on with the double amputee sprinter, he broke down many doors for ILEA. We now see many amputee sprinters who are able to run faster and jump longer than able-bodied athletes. However, the phenomenon could put the International Association of Athletics Federations (IAAF), which governs able-bodied athletics, in an interesting quandary regarding whether the RSPs provide potential advantages or disadvantages for athletic ILEA compared to able-bodied counterparts in running. According to previous studies7,8), top speed during constant- speed running is the product of step frequency, contact length, and the average mass-specific force applied to the running surface during the foot-ground contact period. Therefore, Weyand and Bundle9) advocated that there were three mechanical variables which constrained the speed of human runners: 1) how quickly the limbs can be repositioned for successive steps (shorter swing time), 2) the forward distance the body travels while the foot is in contact with the ground (longer contact length), and 3) how much force the limbs can apply to the ground in relation to the body’s weight (greater body-mass-specific vertical ground reaction force). Weyand and Bundle9) also insisted that if one or more of these variables could be improved, running speeds would be enhanced.

ー 107 ー バイオメカニズム学会誌，Vol. 38，No.2（2014）

6. SWING TIME, CONTACT LENGTH, AND GROUND REACTION FORCES Prosthetic foot inertial properties significantly impact the swing time and stride frequency in running. It has been shown that RSPs were characterized as having lower mass and smaller moment of inertia (MOIs) compared with intact human shank-foot segments10,11). These characteristics may have allowed unnaturally fast leg swing times in the double-transtibial amputee runner using Figure 3 Spring-mass model for hopping (A) and running (B). This model the RSPs. In fact, Weyand and Bundle9) reported that a double- consists of a body mass and a massless linear spring supporting transtibial amputee sprinter in their study showed 21[%] shorter the body mass. The model is shown at the beginning of the swing time and 15[%] higher stride frequency than able-bodied ground contact phase (left), the middle of ground contact phase sprinters during maximal sprint running. In addition, the swing (middle), and at the end of ground contact phase (right). Maximal vertical displacement of the center of the mass and leg spring time of the double-transtibial amputeesprinters was 17.4[%] compression during ground contact is represented by Δy and ΔL, shorter than the first two finishers in the 100[m] dash at the 1987 respectively. Half of the angle swept by the leg spring during the World Track and Field Championships. However, Grabowski ground contact is denoted by θ. et al.12) suggested that the low mass and inertia of RSPs did not facilitate unnaturally fast leg swing times, and that fast leg swing times could result from learning and/or training. 7. LEG STIFFNESS DURING HOPPING AND RUNNING Contact length is defined as the forward distance the body During hopping, jumping, and running, our legs exhibit travels while the foot is in contact with the ground8). Weyand et characteristics similar to those of a spring17). Thus, in these al.13) compared contact length during treadmill running between movements, the musculoskeletal structure of the legs is often one double-transtibial amputee runner and able-bodied runners. modeled with a one-dimensional spring-mass model (Figure 3-A), The authors found that the contact length (normalized to leg which consists of a body mass and a linear leg spring supporting length) of the double-transtibial amputee runner was 9.6[%] the body mass17,18). The leg spring is compressed during the first greater than that of the track athletes, indicating that the greater half of the stance phase and rebounds during the second half. The contact lengths at top speed would also be advantageous for speed. stiffness of the leg spring or leg stiffness (Kleg, defined as the ratio On the other , Grabowski et al.12) compared the contact length of maximal ground reaction force to maximum leg compression at between intact and prosthetic legs in six unilateral transtibial the middle of the stance phase) has been shown to correlate with runners using a treadmill. The authors found that the subjects sprint ability in sprinters, and handball and tennis players19-21). in their study increased the contact length at faster speeds for Thus, identifying Kleg and its relation to sprint performance both, the intact and prosthetic leg, but there were no significant in athletic ILEA would be helpful for determining potential differences in the contact length between the legs. Although these advantages or disadvantages for athletic ILEA as compared with studies examined different amputation types (e.g. bilateral or able-bodied counterparts in running. 22) unilateral amputee runners), the results for contact length were In a previous study , we compared Kleg of sound and prosthetic inconclusive. legs in five runners with lower-extremity amputation using one- Vertical ground reaction force (vGRF) is also an important legged hopping at 2.2[Hz]. We found that there was no significant 8) 11) factor for determining top running speed . Brüggemann et al. difference in Kleg between the legs. Furthermore, based on a compared the vGRF between one double-transtibial athlete and two-dimensional spring-mass model (Figure 3-B), our recent five able-bodied sprinters in maximal overground sprint running study also showed that the prosthetic legs of runners with lower- over 9.2[m/s]. They found that the vGRF was significantly extremity amputation using RSPs has lower Kleg than sound legs higher in the able-bodied athletes than in the double amputee during overground running at 2.5 to 3.5[m/s]15). Our data was sprinter. Similarly, several studies reported that the RSPs would similar with a past finding which demonstrated that there were 12,14-16) impair force generation and thus likely limit top speeds . no distinct differences in Kleg between the legs in ILEA during Unlike swing time and contact length, the results of these studies unilateral running on a treadmill across a variety of speeds16). commonly show that RSPs limits vGRF production during In addition, Wilson et al.23) investigated the effect of prosthetic running, that is, amputee sprinters may have a disadvantage at height and stiffness category on Kleg in two runners with lower- least in force production during running. extremity amputation. They found that these interventions changed

Kleg in the sound leg, but not in the prosthetic leg. In conjunction, these results indicate that RSPs in runners with lower-extremity

ー 108 ー バイオメカニズム学会誌，Vol. 38，No.2（2014）

amputation do not have any potential advantages over biological ground up, Journal of Applied Physiology, 108, 950-961, legs, at least in the spring-like leg behavior. (2010). 8) Weyand, P. G., Sternlight, D. B., Bellizzi, M. J. and Wright, S.: Faster top running speeds are achieved with greater ground 8. CONCLUSIONS forces not more rapid leg movements, Journal of Applied Does this technology enhance performance, or is it essential Physiology, 81, 1991-1999, (2000). for performance?24) The shorter swing times and longer contact 9) Weyand, P. G. and Bundle, M.W.: Point: Artificial limbs do lengths can be viewed as advantages, but there are other potential make artificially fast running speeds possible, Journal of disadvantages to the prostheses, such as lower GRF generation Applied Physiology, 108, 1011-1012, (2010). and leg stiffness in the legs of ILEA. However, our recent study 10) Baum, B. S., Schultz, M. P., Tian, A., Shefter, B., Wolf, suggests that the intact limb in ILEA may be exposed to a greater E. J., Kwon, H.J. and Shim, J. K.: Amputee locomotion: risk of running-related injury than the prosthetic limb or able- Determining the inertial properties of running-specific 25) bodied limbs . As mentioned earlier in this article, the advent of prostheses, Archives of Physical and Rehabilitation, RSPs raised a debate in the scientific community regarding the 94, 1776-1783, (2013). potential advantages or disadvantages of RSPs for ILEA compared 11) Brüggemann G. P., Arampatzis, A., Emrich, F. and Potthast, to able-bodied counterparts in running9,3,26-36). Biomechanists and W.: Biomechanics of double transtibial amputee sprinting physiologists may never be able to quantify all the advantages using dedicated sprinting prostheses, Sports Technology, 1, and disadvantages of running using RSPs. However, one thing 220-227, (2008). is certain; there is a shortage of research involving RSPs. Future 12) Grabowski, A. M., McGowan, C. P., McDermott, W. J., Beale, studies should examine the mechanical characteristics of RSPs M. T., Kram, R. and Herr, H. M.: Running-specific prostheses during running in ILEA and much more research is needed to limit ground-force during sprinting, Biology Letters, 6, 201- provide insights into the potential advantages/disadvantages of 204, (2010). running with RSPs. 13) Weyand, P. G., Bundle, M. W., McGowan, C. P., Grabowski, A., Brown, M. B., Kram, R. and Herr, H.: The fastest runner on artificial legs: Different limbs, similar function? Journal of REFERENCES Applied Physiology, 107, 903-911, (2009). 1) Hobara, H., Baum, B. S., Kwon, H. J. and Shim, J. K.: 14) Baum, B.S., Tian, A., Schultz, M. P., Hobara, H., Linberg, A., Running mechanics in amputee runners using running-specific Wolf, E. J., Shim, J. K.: Ground reaction force and temporal- prostheses, Japanese Journal of Biomechanics in Sports and spatial adaptations to running velocity when wearing running- Exercise, 17, 53-61, (2013). specific prostheses, Proceedings of the 36th Annual Meeting 2) Nolan, L.: Carbon fibre prostheses and running in amputees: a of the American Society of Biomechanics, (2012). review, Foot and Ankle , 14, 125-129, (2008). 15) Hobara H., Baum, B. S., Kwon, H. J., Miller, R. H., Ogata, T., 3) Aruin, A. S.: Sports after amputation, Biomechanics in Sport- Kim, Y. H. and Shim, J. K.: Amputee Locomotion: Spring-like performance enhancement and injury prevention, 637-650, leg behavior and stiffness regulation using running-specific (2000). prostheses, Journal of Biomechanics, 46, 2483-2489, (2013). 4) Scholz, M. S., Blanchfield, J. P., Bloom, L. D., Coburn, B. 16) McGowan, C. P., Grabowski, A. M., McDermott, W. J., Herr, H., Elkington, M., Fuller, J. D., Gilbert, M. E., Muflahi, S. H.M. and Kram, R.: Leg stiffness of sprinters using running- A., Pernice, M. F., Rae, S. I., Trevarthen, J A. and White, S. specific prostheses, Journal of Royal Society Interface, 9, C.: The use of composite materials in modern orthopaedic 1975-1982, (2012). medicine and prosthetic devices: A review, Composites 17) Blickhan, R.: The spring-mass model for running and Science and Technology, 71, 1791-1803, (2011). hopping, Journal of Biomechanics, 22, 1217-1227, (1989). 5) Pailler, D., Sautreuil, P., Piera, J. B., Genty, M. and Goujon, 18) Butler, R. J., Crowell, H.P. and Davis, I. M.: Lower extremity H.: Evolution in prostheses for sprinters with lower-limb stiffness: Implication for performance and injury. Clinical amputation, Annales de Réadaptation et de Médecine Biomechanics, 18, 511-517, (2003). Physique, 47, 374-381, (2004). 19) Bret, C., Rahmani, A., Dufour, A. B., Messonnier, L. and 6) Webster, J. B., Levy, C. E., Bryant, P. R. and Prusakowski, Lacour, J. R.: Leg strength and stiffness as ability factors P. E.: Sports and recreation for persons with limb deficiency, in 100 m sprint running, Journal of Sports Medicine and Archives of Physical medicine and Rehabilitation, 82, S38-44, Physical Fitness, 42, 274-281, (2002). (2001). 20) Chelly, S. M. and Denis, C.: Leg power and hopping stiffness: 7) Weyand, P. G., Sandell, R. F., Prime, D. N. and Bundle, M.W.: relationship with sprint running performance, Medicine and The biological limits to running speed are imposed from the Science in Sports and Exercise, 33, 326-333, (2001).

ー 109 ー バイオメカニズム学会誌，Vol. 38，No.2（2014）

21) Durand, S., Ripamonti, M., Beaune, B. and Rahmani, A.: Leg B. and Herr, H. M.: Counterpoint: Artificial legs do not make ability factors in tennis players, International Journal of Sports artificially fast running speeds possible, Journal of Applied Medicine, 31, 882-886, (2010). Physiology, 108, 1012-1014, (2010). 22) Hobara, H., Tominaga, S., Umezawa, S., Iwashita, K., Okino, 32) Kram, R., Grabowski, A. M., McGowan, C. P., Brown, M. A., Saito, T., Usui, F. and Ogata, T. : Leg stiffness and B., McDermott, W. J., Beale, M. T. and Herr, H. M.: Last sprint ability in amputee sprinters, Prosthetics and Word on Point:Counterpoint: Artificial limbs do/do not make International, 36, 312-317, (2012). artificially fast running speeds possible, Journal of Applied 23) Wilson, J. R., Asfour, S., Abdelrahman, K. Z. and Gailey, R.: Physiology, 108, 1020, (2010). A new methodology to measure the running biomechanics of 33) Lechler, K. and Lilja, M.: Lower extremity leg amputation: an amputees, Prosthetics and Orthotocs International, 33, 218- advantage in running? Sports Technology, 1, 229-234, (2008). 229, (2009). 34) Morin, J. B.: A narrow focus on swing time and vertical 24) Burkett, B.: Technology in Paralympic sport: performance ground reaction force, Journal of Applied Physiology, 108, enhancement or essential for performance? British Journal of 1017-1018, (2010). Sports Medicine, 44, 215-220, (2010). 35) Weyand, P. G. and Bundle, M. W.: Last word on 25) Hobara, H., Baum, B. S., Kwon, H. J., Linberg, A., Wolf, E., point:Counterpoint: Artificial limbs do make artificially fast Miller, R. H. and Shim, J. K.: Amputee Locomotion: lower running speeds possible, Journal of Applied Physiology, 108, extremity loading using running-specific prostheses, Gait and 1019, (2010). Posture, 39, 386-390, (2014). 36) Zelik, K. E.: "Net advantage" is more rooted in sport than 26) Adamczyk, P. G.: For forward running, study fore-aft forces, science, Journal of Applied Physiology, 108, 1016-1017, Journal of Applied Physiology, 108, 1017, (2010). (2010). 27) Buckley, J. G. and Juniper, M. P.: Artificial limbs can enable artificially fast running, Journal of Applied Physiology 108, 1016, (2010). 保原浩明（ほばら ひろあき） 28) Cavagna, G. A.: At high running speeds, power developed 2008 年早稲田大学大学院人間科学研究 each step during the push appears to be sustained by elastic 科博士後期課程修了．博士（人間科学）． energy, Journal of Applied Physiology, 108, 1016, (2010). 国立障害者リハビリテーションセンター 29) Chockalingam, N., Thomas, N. B., Smith, A. and Dunnin, 研究所流動研究員，日本学術振興会特 D.: By designing 'blades' for Oscar Pistorius are prosthetists 別研究員（PD）， University of Maryland, creating an unfair advantage for Pistorius and an uneven Research Associate を経て，2013 年より（独）産業技術総合 playing field? Prosthetics and Orthotics International, 35, 482- 研究所デジタルヒューマン工学研究センター研究員．スポー 483, (2011). ツ用義足の生体力学的評価に関する研究に従事．国際バイ 30) Fuss, F. K.: Closing the gap through technology, Sports オメカニクス学会，日本バイオメカニクス学会などの会員． Technology, 1, 169-171, (2008). 31) Kram, R., Grabowski, A. M., McGowan, C. P., Brown, M.

ー 110 ー