September 2011

Effects of load carriage on amputee ambulation #15456961

Amputees lack the lower extremity muscles responsible for maintaining kinematic stability under increasing load carriage conditions, and would benefit from a more versatile prosthetic foot design that could adapt dynam­ically to changing loads.

By Kurt Collier, CP

A thorough understanding of the body’s typical performance in the absence of limb loss facilitates the clinician’s task of restoring function following amputation. The prosthetic rehabilitation team has long benefited from a basic appreciation of human gait, including its force patterns, joint movements, and sequencing events. As practitioners, however, we are largely ignorant of the body’s typical response to load carriage, or the bearing of additional weight during standing and ambulation.

This deficit is unfortunate as many of the vocational and avocational tasks encountered by the amputee community involve the lifting and carriage of additional burdens, and these augmented loads affect the stability, responsiveness, and safety of prosthetic systems. Fortunately, this area has received extensive investigation in both military and civilian settings, and these findings describe a rather consistent pattern of the body’s response to augmented loads and provide a clearer understanding of prosthetic implications.

Able-bodied response to load carriage

Figure 1. Double bump pattern of the vertical ground reaction force, increasing at an approximate 1:1 ratio with increasing theoretical load carriage conditions. (Derived and adapted from references 2, 4, and 5. This figure is intended to be theoretical and illustrative, therefore force values are not provided.)

The vertical ground reaction forces observed during normal human locomotion are characterized by a vertical “double bump” pattern, with increasing force observed during both the weight acceptance component of the loading response phase and the “roll-off” component of terminal stance phase.1 Additional horizontal forces are experienced in the sagittal plane in the form of early, negative braking forces and later, positive propulsive forces.

Observations of both vertical and horizontal ground reaction forces in the presence of augmented load carriage rates of 20% to 66% of body weight have confirmed that all forces undergo proportional increases in the presence of additional loads (Figures 1 and 2).2-5 For the vertical forces, this increase is generally directly proportional to the amount of augmented load carriage, with an extra pound of force throughout the gait cycle for every extra pound of augmented load. The sagittal horizontal forces also increase, although at a reduced ratio.2-5 The vertical and horizontal ground reaction forces are affected at different points in the gait cycle. The peak negative horizontal force is experienced during loading response, the peak positive horizontal force is experienced during late stance, and the peak vertical ground reaction force begins at midstance and increases through terminal stance.

These ground reaction forces occur across the various joints of the lower extremity, with those at the ankle being the most relevant to prosthetic foot considerations and their influences at proximal joints. At the beginning of each step, these external forces act on the ankle to promote the plantar flexion observed during loading response. During the end of each step, as the ground reaction force passes increasingly anterior to the ankle joint, an external moment is created that would drive the ankle into dorsiflexion.1

As might be expected, as the magnitude of the ground reaction forces increase with load carriage, concomitant increases are observed in the external moments that must be countered at the ankle joint. The increases in these external moments during loading response are minimal, however, due to both their brief duration and the short length of the heel lever. These moments become much more substantial during mid- and terminal stance, with increases in the external ankle moments of 20%, 27%, and 38% corresponding respectively to augmented load carriage conditions of +26%, +43%, and +61% of body weight (Figure 3).2,3,5

Figure 2. Horizontal ground reaction forces about the ankle in the sagittal plane, characterized by a negative “braking” force followed by a positive “propulsive” force. Progressively larger positive and negative force magnitudes are experienced with augmented theoretical load carriage conditions. (Derived and adapted from references 2, 4, and 5. This figure is intended to be theoretical and illustrative, therefore force values are not provided.)

Given the observed increases in the external joint moments acting across the ankle joint during load carriage, it would seem reasonable to suspect an alteration in observed ankle kinematics. Specifically, increased or excessive dorsiflexion during terminal stance would seem likely in the presence of augmented vertical ground reaction forces associated with load carriage conditions. The external joint moments at the ankle are what initiate joint motion and necessitate resistive forces at the ankle joint. Any increase in these moments must result in either augmented ankle motion or augmented resistance at the ankle. However, numerous investigations in both military and civilian settings have consistently reaffirmed that sagittal ankle kinematics throughout the gait cycle are essentially unchanged across a broad range of load carriage conditions.2,3,6-9

The ability of the lower extremities to preserve similar joint kinematics at the ankle under variable load carriage conditions suggests the ability to generate a range of internal joint moments to counter the elevated external moments that result from the larger ground reaction forces. The gastrocnemius, soleus, and, to a much lesser extent, the tibialis anterior muscles have all been seen to generate increased activity during load carriage.

For example, the activity of the gastrocnemius reportedly increases by 11% and 34%, respectively, with load carriage rates of +26% and +43% of bodyweight.2 Increases in soleus activity are even more striking, reaching roughly 50% under load carriage conditions of +50% of body weight.5 In contrast, the need for augmented internal resistance during the loading response phase of gait appears minimal. Tibialis anterior activity during the same load carriage conditions of +26% and +43% of body weight increased by only 4% and 6%, respectively.2

Figure 3. External moments about the ankle: A brief negative dorsiflexor moment at the beginning of stance is followed by a much larger positive plantar flexor moment that increases throughout mid- and terminal stance. Although the dorsiflexor moment is minimally affected by theoretical augmented load carriage conditions, the plantar flexor moments increase with theoretical increases in load carriage. (Derived and adapted from references 2 and 5. This figure is intended to be theoretical and illustrative, therefore moment values are not provided.)

Increases in load carriage create augmented vertical and horizontal ground reaction forces. These forces increase the external moments acting across the joints of the lower extremity, including the ankle. However, even in cases in which augmented load carriage conditions exceed 50% of body weight, observed ankle kinematics remain largely unchanged. This is accomplished through increased activity of the muscles of the lower extremity that generates sufficient internal joint moments to counter the effects of the augmented ground reaction forces. In the case of the plantar flexors during late stance, this augmented muscle activity appears to be quite substantial.

Prosthetic findings and implications

The exact burden of load carriage within the amputee population is only minimally understood. Roy et al reported on a small sample of six individuals with transtibial amputations who were asked to ambulate under load carriage conditions of +11%, +22%, +33%, and +44% of the group’s mean body weight.10 Investigators collected energy expenditure data using analyses of expired air and compared the information against data gathered from matched controls.

Amputees experienced elevated energy expenditure, oxygen consumption, and heart rate for the no-load condition compared to controls—differences that were exacerbated by load carriage. The authors reported 70% greater increases in energy expenditure in the amputees compared with controls for the first three load carriage conditions. For the most aggressive load carriage condition of +44% of body weight, this relative increase in energy expenditure reached 82%.10 Increases in oxygen consumption values were also, on average, 50% to 65% greater than those observed in controls, with more pronounced increases observed under higher load carriage conditions. Finally, increases in peak heart rates were between 16% and 22% greater for the amputee group, with the differences becoming more striking as load carriage increased.10 However, the mechanisms behind these increases in energy expenditure, oxygen consumption, and heart rate within the amputee group were not explored at any depth in this early publication.

Figure 4. Dual-keel technology in body weight only and augmented load carriage conditions. Top images show schematic illustrations; bottom images show artistic renderings of the technology. a) Body weight only: in the absence of augmented load carriage, only the primary keel is engaged during late stance. b) Augmented load carriage: increasing ankle moments cause deflection of the primary keel and engage the secondary keel.

Possible mechanisms underlying these metabolic changes with load carriage in the amputee population have been evaluated in conjunction with a military prototype foot that uses a robotic tendon to mimic action of the plantar flexors.11 Unlike the studies of load carriage in nonamputees, where external joint moment values were derived in the gait laboratory through inverse dynamics, the prosthetic tendon allowed for the direct measurement of internal resistive joint moments. Specifically, knowing the stiffness of the spring represented by the tendon and the range of deflection experienced during a given period of gait, Hooke’s law could be used to directly calculate the moments acting about the ankle joint.11

As part of the preliminary investigations for a case study, researchers had a transtibial amputee walk under a range of load carriage conditions (body weight only, body weight +4.5 kg, and body weight +9 kg) and under three different posterior spring conditions of progressing stiffness. Investigators found noticeable, progressive increases in the peak ankle moments under augmented load carriage conditions across all three spring conditions. Even though these load carriage conditions only represented increases of 5.5% and 11% of body weight, peak ankle moments increased by 3% and 14%, respectively, indicating a statistically significant interaction between load and recorded ankle moments.11

In addition to affirming the effect of load carriage on joint moments in a prosthetic condition, the investigators were able to evaluate the effect of the spring stiffness on the resulting moments about the ankle and the respective metabolic implications. Spring stiffness represents the prosthetic foot’s resistance to dorsiflexion, or the ability of the foot to create internal moments to counter the external moments from ground reaction forces acting across the prosthesis.

In the case study in question, the authors collected data on “input energy,” or the actual energy usage by the motor of the prosthetic tendon across the various conditions described above. These data were compared against a predictive power model based on the mechanical prosperities of the robotic tendon system. Observing good correlation between the collected data and model predictions, the authors verified that the energy costs of ambulation increase with both the augmented load carriage and the stiffness of the spring (or the prosthetic foot’s resistance to dorsiflexion).

Importantly, according to this model, when load carriage exceeds about 20% of body weight, the input energy of ambulation decreases with stiffer spring values.11 Put more pragmatically, the ideal stiffness of a prosthetic foot varies according to the loads being carried. Once load carriage values exceed 20% of body weight the energy costs of ambulation will rise if the stiffness of the prosthetic system does not increase.

Existing and developing solutions

In the rapidly developing world of prosthetic refinements, mechanisms of variable resistance to dorsiflexion are likely to become increasingly elegant and sophisticated. A dual-keel design represents an attempt to provide this function in a passive manner (Figure 4). In this design, the heel plate and primary keel on the most plantar aspect of the foot are designed with a predetermined stiffness to provide appropriate resistance to both plantar flexion and dorsiflexion in the unloaded condition.

Figure 5. Comparative foot stiffness under augmented load carriage conditions. Both feet follow similar deflection curves during normal load conditions. However, under augmented load carriage the single-keel design experiences greater deflection while the stiffness of the dual-keel design is more resistant. (Source: Unpublished in-house testing data, Freedom Innovations.)

In conditions involving load carriage, the heel plate deflects until it contacts a more proximal heel-stiffening bumper, mimicking the increased action of the tibialis anterior during load carriage. As the step proceeds and weight is transferred to the toe of the foot, the primary keel deforms under the additional load, engaging the secondary upper keel and providing augmented dorsiflexion resistance, mimicking the increased activity of the plantar flexors during load carriage (Figure 5).

Meanwhile, researchers are pursuing more elaborate, dynamic solutions. The robotic tendon concept cited earlier has been suggested as a means of creating a powered prosthetic foot capable of dynamically modulating spring stiffness as a function of loading to optimize energy efficiency.11


The body’s normal response to augmented load carriage is both well studied and consistently observed. Despite the resulting variance in external joint moments, which increase in response to augmented ground reaction forces, the body retains a model of kinematic invariance. This relies on the ability of key muscle groups, such as those in the gastroc-soleus complex, to increase their activity and offset elevated external joint moments.

The constant stiffness characteristics of current prosthetic foot options represent a substantial limitation in their functionality.12 In the absence of variable resistance to dorsiflexion, amputees who encounter variable load carriage requirements throughout the day must adapt by either interchanging between several prosthetic feet or using a single foot that is likely suboptimal for multiple environments.12

Given the frequency with which amputees must function under conditions of augmented load carriage, the potential benefits of a more versatile prosthetic foot design are readily apparent. Although research is underway to provide prosthetic solutions capable of dynamic adaptation to load carriage, the option of a dual keel in a passive prosthetic foot may provide a mechanism of variable resistance to better adapt to load carriage conditions.

Kurt Collier, CP, is the vice president of clinical services for Freedom Innovations in Irvine, CA.


  1. Perry J, Burnfield JM. Gait Analysis: Normal and Pathologic Function. 2nd ed. Thorofare, NJ: Slack; 2010.
  2. Harmen EA, Hoon K, Frykman P, Pandorf C. The effects of backpack weight on the biomechanics of load carriage. Technical Report T00/17, MA: United States Army Research Institute of Environmental Medicine; 2000:1-71.
  3. Polcyn AF, Bensel CK, Harmen EA, et al. Effects of weight carried by soldiers: combined analysis of four studies on maximal performance, physiology and biomechanics. Technical Report TR-02/010, MA: United States Army Research Institute of Environmental Medicine; 2002:1-64.
  4. Birrell SA, Hooper RH, Haslam RA. The effect of military load carriage on ground reaction forces. Gait Posture 2007;26(4):611-614.
  5. McGowan CP, Neptune RR, Kram R. Independent effects of weight and mass on plantar flexor activity during walking: implications for their contributions to body support and forward propulsion. J Appl Physiol 2008;105(2):486-494.
  6. Attwells RL, Birrell SA, Hooper RH, Mansfield NJ. Influence of carrying heavy loads on soldiers’ posture, movements and gait. Ergonomics 2006;49(14):1527-1537.
  7. Lee M, Roan M, Smith B. An application of principle component analysis for lower body kinematics between loaded and unloaded walking. J Biomech 2009;42(14):2226-2230.
  8. Birrell SA, Haslam RA. The effect of load distribution within military load carriage systems on the kinetics of human gait. Appl Ergon 2010;41(4):585-590.
  9. Smith B, Roan M, Lee M. The effect of evenly distributed load carrying on lower body gait dynamics for normal weight and overweight subjects. Gait Posture 2010;32(2):176-180.
  10. Roy AK, Ganguli S, Datta SR. Performance evaluation of BK amputees through graded load carrying test. Acta Orthop Scand 1977;48(6):691-695.
  11. Hitt J, Sugar T. Load carriage effects on a robotic transtibial prosthesis. Presented at: 2010 International Congress on Control, Automation, and Systems, Gyeonggi-do, Korea, October 2010.
  12. Hitt J, Merlo J, Johnston J. Bionic running for unilateral transtibial military amputees. Presented at: 27th Army Science Conference, Orlando, November 2010.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.