Disabilities have been included in snow sports for many years, and those with limb absences and wheelchair users are the most recognizable skiers with disabilities on the slopes.1 Prostheses are fundamental for daily living and to allow participation in sport.2
Everyday prostheses are generally inadequate for skiing.3 There is limited componentry available for high-activity sports. Understanding the biomechanics involved in bipedal (able-bodied) and two-track skiing are essential to allow the user to function optimally. Two-track skiing involves skiing with two skis with the aid of a lower-limb prosthetic device and no other aids.4,5
The aim of this study is to compare the kinematics of a bipedal skier with two-track skiers using a transfemoral prosthesis while turning in both directions using a video analysis system (Siliconcoach, Dunedin, Otago). The COM position was calculated, and the joint angles were measured and compared for different phases of a ski turn.
LITERATURE REVIEW METHODOLOGY
Three literature searches were conducted: 1) skiing biomechanics, 2) prosthetic skiing, and 3) prosthetic skiing components. Searches were conducted in: Medline, ProQuest Engineering Research Database, ProQuest ABI/Inform Complete, Cochrane library, Scopus, SportDISCUS, Science Direct, Google Scholar, and Google. Searches were conducted using the terms: “alpine skiing or downhill skiing” and “biomechanic* or kinematic*” for skiing biomechanics, “alpine skiing or downhill skiing” and “prosthe* or artificial limb or limb loss or amput*” for prosthetic skiing, and “skiing or high activity or Snowsport*” and “artificial limb* or prosthetic knee* or prosthetic ankle*” for prosthetic skiing components. Inclusion criteria were applied, and irrelevant studies were removed. Studies reporting on aged, elderly or preschool skiers, and those associated with injuries, surgery, or prosthetic joint replacement were excluded.
LITERATURE REVIEW RESULTS
In all three literature reviews, studies were assessed using SIGN [Scottish Intercollegiate Guidelines Network] Guidelines6 and Websites using CARS [Credibility, Accuracy, Reasonableness, Support] checklist.7
In the first search for skiing biomechanics, nine studies remained after the inclusion-and-exclusion criteria were applied, and the top five articles, all rated high based on relevance and methodology when assessed, were selected for inclusion.8–12
The second literature search found very few articles relating to the use of prostheses while skiing; therefore, some articles that described all types of skiing available for persons with lower-limb absence were included.4,5 After inclusion and exclusion criteria and assessment, four studies1,2,5,13 and one Website4 were selected.
In the third search for prosthetic skiing components, only two relevant articles3,14 were assessed and found to be of high quality, based on relevance and methodology. Two news articles15,16 of lower quality were also included. The manufacturers' information for components mentioned in the literature was included where possible; therefore, five manufacturer's Websites17–21 and four articles3,14–16 were included.
LITERATURE REVIEW DISCUSSION
Biomechanics of skiing techniques are often explored because the ski turn is fundamental in controlling speed.8 The basic phases are the same for the two main techniques: skidding and carving turns.9,10 At the point where one turn ends and the next begins, the skier is almost perpendicular to the fall line.11 The fall line is the point where the skier faces straight down the slope. Ski turns can be divided into three phases: the initiation phase, steering phase I, and steering phase II.8,12
The initiation phase is characterized by a load change and an edge change.9,12 This is achieved through unweighting, which is essential to being able to turn the skis.9,12 Unweighting is a reduction in force on the slope, which is achieved through hip and knee extension.9,12 The outside knee starts at 68° to 85° of flexion and extends to 45° to 60° of flexion.9,12 Most of the movement comes from the knees (which both extend to 45°–60° of flexion) and hips.12
The steering phase is achieved once the skis start to turn.8 Steering phase I is rotation into the fall line, and steering phase II is rotation out of the fall line.8,9,12 Rotation and loading the inside edge of the outside ski will turn the ski.12 To maintain balance, the skier has to lean at the correct angle toward the center of the rotation. In steering phase I, the lateral force acts toward the center of the turn, so small edging angles and inward leaning are required.12 The outer leg is flexed 50° to 70°, while the inner is flexed 60° to 68°.9,12 In steering phase II, the lateral force acts away from the center of the turn, so larger edging angles are then required to maintain balance, which results in greater knee flexion of the inner leg (68°–85°) than the outer leg (50°–70°).9,12 To prevent excessive slippage, the shoulders, hips, and knees need to be in a neutral position to distribute the ground reaction force evenly along the downhill ski.9,12
Although the basic phases are the same for carving and skidding turns, there are a few differences. In carving, there is a more uniform force applied to the outer ski, whereas for skidding, the forces are applied to the front part of the ski.9,10 In summary, the knee flexion angles for bipedal skiers are:
- Outside knee
- Initiation, 45° to 60°
- Steering phase I and II, 50° to 70°
- Inside knee
- Initiation, 45° to 60°
- Steering phase I, 60° to 68°
- Steering phase II, 68° to 85°
There are a number of techniques available for standing skiers with amputation.4,5,13 Four-track skiing uses two skis and two outriggers,4,5,13 which are forearm crutches where the end of the crutch has a very small ski on it so it slides on the snow.1,5 Two-track skiing is used by those who do not need outriggers; persons with below-knee amputations often use this method with the aid of prostheses.4,5 A three-track skiing technique, using one ski and two outriggers, is typically used by those with unilateral transfemoral amputations because their prostheses are difficult to control.1,4,5,13 In addition to three-track skiing, a few varieties of prosthetic knee and ankle components have been developed in recent years for regular two-track skiing.14
Skiing requires more dorsiflexion than walking; therefore, everyday prostheses are often inadequate.3 The Slalom Ski Foot (Freedom Innovations, Irvine, CA) is a prosthetic foot designed specifically for skiing, which inserts directly into the ski binding, increasing the skier's control and eliminating the need for a boot.17 There are also prosthetic knees available, which are suitable for skiing. The C-Leg and Genium (Ottobock global, Salt Lake City, UT) allow weight bearing in a position of flexion.18,19 They are microprocessor knees designed for walking with additional activity modes, which allow skiing.18,19 They are only splash resistant and are not designed specifically for high activity; because snow is wet and skiing is high impact, this may damage these expensive components.19 The Ottobock X3 microprocessor knee is completely waterproof.20
The XT9 (K12 Prosthetics, Saratoga Springs, UT) is a prosthetic knee that was designed for high-activity sports and can be used for two-track skiing.14 The XT9 energy-storing prosthetic knee mimics the functions of the quadriceps in high-activity sport.21 Flexion is constantly resisted by a progressive energy coil spring and is adjustable for different user weight and activity.21 It is fully waterproof, able to endure extreme climates, and suitable for high-impact sports, including snow sports.21 However, it is not suitable for everyday use.15 When skiing with the XT9, the center of gravity moves backwards on knee flexion.14 This results in the skier losing dynamic balance and control of the ski.14 The Art-Leg Sport Knee was designed with this problem in mind. The Art-Leg allows flexion, extension, and forward translation.14 Forward translation means the knee simulates dorsiflexion of the ankle component with the extra pivot point at its base.14 This occurs during flexion and enables the skier to maintain the optimum position of the center of gravity.14
Three methods for determining the skier's center of mass (COM) were considered to determine the most appropriate method. Body segment parameters (BSPs) look at different characteristic features of the body segments. The mass of each body segment, calculated as a percentage of total body weight, and the COM of each body segment were used to calculate the COM of the body as a whole in different positions.22
The WALKiNSENSE (Nordic Ortopedica) pressure-sensitive insoles (PSIs) determine the center of pressure (COP) along the length of the foot. The laser posture (LP) is used to visualize the position of the COM while standing on a force-sensing platform; a laser projects the COM line onto the body.23
The COM calculated using BSPs and the COP determined by the PSIs were compared with the COM projected by the LP to ascertain an appropriate method for comparison of the COM of different skiers.
PRELIMINARY EXPERIMENT: METHODOLOGY
An able-bodied skier with PSI in the boots stood on the LP device (a clinical tool) while being filmed in the sagittal plane. The COM was calculated using BSP for different skiing positions. The BSP and PSI were compared with the position on the LP. Weight is not placed through the ski poles at any point in the turn, so it has little effect on the COM calculations. The five different skiing positions were as follows:
- Standing, a relaxed position in the sagittal plane
- Fore, leaning forward (hips as far forward as possible)
- Aft, leaning back (hips as far back as possible)
- Flexion, position with maximum hip and knee flexion and ankle dorsiflexion
- Extension, position with maximum hip and knee extension and minimum dorsiflexion
PRELIMINARY EXPERIMENT: RESULTS
Table 1 compares LP, BSP, and PSI. The distance from the tip of the toe (0 cm) to LP and BSP in centimeters is shown in brackets. Table 1 also shows the difference between the LP and BSP in centimeters. For comparison with PSI, LP and BSP results were split into four sections relating to the position along the foot. With a boot length of 29 cm, the four sections are as follows:
- 1. Anterior to the forefoot (<0 cm)
- 2. Forefoot (0–9.7 cm)
- 3. Middle (9.7–19.3 cm)
- 4. Heel (19.3–29 cm)
The PSI data, LP, and BSP for the aft and fore positions are shown in Figure 1. The red and green lines represent the LP and BSP, respectively.
PRELIMINARY EXPERIMENT DISCUSSION
For all the different positions in the sagittal plane, the LP and BSP were very similar with a mean difference of 1.7 ± 0.8 cm (Table 1), and they consistently acted in the same area of the foot (Table 1). This indicates that the BSP calculations are an appropriate method for determining the COM in skiing. The PSI, LP, and BSP all show the most pressure at the heel for the aft position (Figure 1A). The PSI was the same as the LP and BSP with the exception of the fore position (Table 1). In the fore position, LP and BSP are anterior to the forefoot as demonstrated by Table 1 with −7.4 cm for LP and −8.9 cm for BSP, and illustrated by Figure 1B, whereas the PSI are at the forefoot (Figure 1B). This is because the PSI only detects pressures within the boot (i.e., COP) and therefore cannot distinguish between forefoot COM and anterior to forefoot COM (i.e., between sections 1 and 2). This is possible because the skis are much longer than the boots. Therefore, pressure-sensitive equipment would ideally need to cover the whole length of the ski or, at the very least, the length of the binding to be able to show COM and not just COP. The LP is a clinical tool, and the PSIs are limited to the pressures that fall within the ski boot; therefore, the BSP was deemed suitable for use.
MATERIALS AND METHODS
Research was granted ethical approval by the University of Strathclyde ethics committee number UEC13/72.
A bipedal skier and 2 two-track skiers were analyzed while skiing through a set course. Participant 1 (P1) was a bipedal skier; participants 2 and 3 (P2 and P3) were two-track skiers selected at random from a ski school for those with limb absences. They were required to have previous skiing experience and be aged 21 to 65 years. Both P2 and P3 used the XT9 prosthetic knee and SACH foot (1D10) on the right side. Standard pieces of equipment (skis and boots) with no alterations were used.
Measurements required to calculate the COM using BSPs were taken, including height, weight, prosthesis weight, sound side femur length, and circumferences plus residual limb length and circumferences. The COM of the prosthesis in the sagittal plane was determined. The prosthesis was allowed to rotate freely on a pivot point, which prevented transverse motion. This kept the prosthesis in the same plane, which is essential because the method works best for 2D objects. A length of string with a weight on the end was attached to the prosthesis at the pivot point; once both string and prosthesis were at rest, the string was attached to the prosthesis at both ends. This was repeated three times with different pivot points. This resulted in three pieces of string that all crossed over at the same point. The position where the pieces of string met is the position of the COM in the sagittal plane.24 The COM was found to be at knee-joint level on the prosthesis in the sagittal plane.
After discussion with experts in the field and based on the International Ski Federation (FIS) regulations, the course dimensions (Figure 2) were decided and the fifth and sixth turns were filmed.25 A ski pole was positioned 1.5 m in from the stubby (turning marker), and all four cameras were 7.5 m from this ski pole (Figure 2).
The markers were positioned on the participants and included two hip markers, two ankle markers, four rotational indicators, and an elastic strap below the knee. The rotational indicators act as visual aids to help determine the skiers' rotation.26 The large (thigh) and small (ankle) rotational indicators were validated with Siliconcoach at the distance the participants were in the study. One individual positioned the markers in different rotations, and another determined the angle from the video. The mean difference between the angle set and the angle measured was 3° ± 5.7° for the thigh markers and 10.5° ± 11.9° for the ankle markers. The larger markers can be used to determine the rotation of the skier, whereas the smaller indicators should be used with caution.
Visual and audio signals were given before each run. Six runs were completed by all participants with the cameras at turn 5 and then eight runs at turn 6.
The videos were analyzed using video analysis software Siliconcoach. The participants' knee flexion and tibial inclination angles (TIAs) were measured in the sagittal plane in the three phases of the turn for both the left and right turns. The participants' knee flexion angles were compared with that of bipedal skiers. The TIAs and width of the feet were compared between the three participants.
For the COM calculations using BSP, the videos were analyzed and the key frames (n = 84) were selected; this included all participants (n = 3), all phases (n = 3), and all runs (n = 14).22 The rotation indicators were used to aid selection of the key frames, in which the sagittal view was desired. The COM was calculated in the x and y direction to give the height of the COM and the position of the COM in relation to the participants' boots. The COMs for the participants were compared.
For comparison between the participants, the mean COM and the mean joint angles (of the inside and outside legs) of each participant during all three phases of a right and left turn are shown in degrees (Table 2). The angles that fall within the reference range are highlighted in gray. The COM position in the sagittal plane was expressed as a percentage of the length of the boot from the most anterior aspect (tip) of the boot and as a percentage of the height of the body from the ground. Participant 1 and P3 were advanced skiers, and P2 was advanced intermediate.
When compared with the normal knee flexion angles for bipedal skiers, P1 falls within the reference range for all phases for both knees during the right (Table 2A) and left turn (Table 2B). The knee flexion angles at each phase and the change in knee flexion throughout both turns are less for P2 and P3 compared with P1. This suggests possible difficulty flexing the prosthetic knee.
For the right turn, P3's knees were less flexed than normal for steering phase I and II and P2's knees were always less flexed than normal (Table 2A). The highest flexion achieved was 60.8° ± 3.6° using the XT9 knee; this was by P2 in steering phase II (Table 2A). Although this is suitable when the prosthesis is the outside leg, it is not comparable for the inside leg where a minimum of 68° is considered normal.9,12 The reduced flexion of the outside knee for P2 and P3 may be to compensate for the reduced flexion of the prosthetic knee to achieve the correct angle toward the center of rotation.12 The TIAs were higher for the inside leg than the outside for P2, suggesting that the outside limb is stretching to put more weight on the inside (prosthetic) leg to flex the knee when the weight should be on the outside leg.9 Demšar et al.14 found that load on the inside prosthetic leg was due to flexion resistance.
For the left turn, P3 was within the reference range for both legs during initiation and for the outside knee in steering phase I. Participant 2 was only within the reference range for the outside leg during initiation 45.3° ± 10.2° (Table 2B). The highest TIA for the prosthetic leg (using SACH foot 1D10) was 23.6° ± 3.2° for P2 and 23.1° ± 3.6° for P3; however, P1 was able to achieve 30.8° ± 3.2° of tibial inclination (Table 2B). This lack of dorsiflexion prevents forward translation of the knee so the COM is unable to move anteriorly.
The COM for P1 is consistently anterior (COM, <50%) to the middle of the boot, with 44.9% as the most posterior and −0.05% the most anterior, which is just anterior to the tip of the boot (Table 2A). For P3, the COM was always posterior (COM, <50%) to the middle of the boot ranging from 53.0% to 79.3% (Table 2). Participant 2's COM was posterior to the midline of the boot for all phases of the right turn (Table 2A) and posterior to the whole boot during initiation (115.1%). However, it was just anterior to the midline of the boot for steering phase II of the left turn at 45.8% (Table 2B). The height of the COM for P2 was generally higher for the right turn, which further indicates that there was difficulty flexing the prosthetic knee when on the inside.
For P1, the TIA was always the highest and the COM was always the most anterior. This suggests that the TIA is important to position the COM anterior to the middle of the boot, which is required for skidding turns.10 When comparing initiation and steering I for the left turn of P2, the knee flexion angles are very similar (45.3° and 45°, Table 2B). However, TIA was higher for initiation at 21.7° ± 4.4°, compared with 19.6° ± 3.4°, and the COM was more anterior for initiation at 59.4%, compared with 86.7% for steering I, further showing the importance of TIA.
The prosthetic knee joint is also important for COM position. The TIA is the same for steering II of the left turn for P2 and P3; however, the knee angles are different: P2 at 46.7° ± 2.3° and P3 at 49.3° ± 2.8° (Table 2B). The COM for P2 is more anterior (45.8%) compared with P3 (54.0%). This demonstrates that as the prosthetic (outside) knee flexes it causes the COM to move more posteriorly. One possible explanation for the outside (prosthetic) knee on the left turn for P2 and P3 being less flexed than in normal bipedal skiing is because flexion results in posterior COM. Therefore, the weight is not over the whole or front of the ski, which can cause the skier to lose control resulting in slippage of the ski.12
LIMITATION AND FUTURE STUDY
The results may have been affected by the difference in ability of the skiers and should be considered within this limitation. There was also a small sample size in this study, so future studies should use larger, statistically significant sample sizes to look into the issues raised in this study. Prostheses may need to be set up with increased TIA or allow increased dorsiflexion on weight bearing to optimize the skier's position. The prosthetic knee needs to limit flexion through the whole turn, but optimally needs to allow more flexion when little weight is on the prosthesis. These would need to be compared with current prostheses for skiing to determine the effects on the COM position.
The two-track skiers had difficulty flexing the prosthetic knee when on the inside of the turn because there is little weight going through the inside (prosthetic) leg. There was also reduced knee flexion when the prosthetic leg was the outside (weight-bearing) leg. However, it was found that increased knee flexion resulted in posterior displacement of the COM due to a lack of tibial inclination, which could lead to poor ski control. Therefore, translation of the knee anteriorly on flexion and more knee flexion when not weight bearing may improve the kinematics of two-track skiers. The tibial inclination that is affected by the prosthetic foot and setup is also important for achieving similar kinematics to bipedal skiers.
The authors would like to acknowledge the ski instructors interviewed for their skiing expertise and the participants for their time when taking part in the study.