Identification of the forces developed by upper limbs in various forms of human physical activity and in manual techniques used by physiotherapists – a brief review

In recent years, Nordic Walking has become very popular, especially among the elderly. This complex, pro-health physical activity, used as a supplementary exercise by the Scandinavian ski runners already in the 1920s, is recognized as a sport for all. Complex benefits obtained by people in different age groups and at different levels of physical fitness resulting from taking up Nordic Walking regularly were broadly described in the literature [1–6]. Many authors [1,3] measured different variables characterizing physical exertion while walking with poles. Energy expenditure determined with the indirect calorimetry method, based on the maximal oxygen intake, heart rate, blood lactate or hemodynamic parameters was the most frequently analysed value [1]. The influence of Nordic Walking on the body of the elderly [after 1], including functional fitness [7] or body composition [6], was analysed as well.

Many authors underline positive benefits of Nordic Walking in the rehabilitation of patients with acute coronary syndrome, Parkinson’s disease [8], intermittent claudication [9], coronary arteries disease, after heart attack [1,2], with chronic pain in the lumbar spine [5] or other chronic diseases [2]. Nordic Walking is recommended to children, youth, the elderly as well as to individuals with such health problems as arterial hypertension, arteriosclerosis, arthritis, the so-called low back pain, sciatica, osteoporosis, depression or obesity [1].

The effects of the rehabilitation with the use of Nordic Walking were evaluated most frequently on the basis of changes in the results of walking tests of different duration, including the use of space-time walk analysis, maximal oxygen intake, ECG, range of motion in joints, results in physical exercises or functional tests (e.g. Senior Fitness Test), as well as results of life quality assessment, experiencing physical effort or pain [1]. Measurements were also made to evaluate physical stress exerted on people performing this form of activity. For example, Hartvigsen et al. [5] measured physical activity during walking with poles with the use of the accelerometric method, more specifically, using an accelerometer Actigraph (GT256), which was placed around the waist of the subject and was registering vertical movements of the body mass center. The work performed was defined as a total distance covered by the body mass center, and the intensity as an amplitude.

From biomechanical perspective, activity during Nordic Walking was evaluated most frequently with the use of a videographic method (three-dimensional analysis of movement) and registration of the ground reaction force exerted on a force plate, by analysing internal (muscular) values of the moments of force and angles in hip, knee or talocrural joints, or ground reaction forces. For example, Wilson et al. [10], while analysing healthy men and women with the use of the videographic method (three-dimensional analysis of movement, measurements with the inverse dynamics method), showed that average values of ground reaction force and the force in a knee joint while walking with poles were lower than while walking without poles.

In order to determine values of forces acting along the longitudinal axes of the poles, Schiffer et al. [11] used special measuring poles with integrated force sensors. The authors, who surveyed 13 young women (Nordic Walking instructors) showed that the values of force-time parameters pointed to considerable inter-individual differences and depended not only on the techniques of walking with poles, but also on the type of road surface. Average values provided by the authors [11] were as follows: axial forces from 36.5 to 43.3 N, force impulses transferred to the ground with the poles – ca. 8 N s, contact time of the poles with the ground – ca. 0.4 s, maximal force rate from 500 to 700 N/s, whereby at the first stage of the contact between the pole and the ground, maximal force rate exceeded 1200 N/s. Pšurný et al. [12] examined upper limb axial forces on a pole during the support phase of 17 individuals (10 men and 7 women at the age of 25.9±3.6 years) by MPAF system from the Czech Republic. The sensor was inserted by cutting a pole. Researchers were taking two factors into account: inclination angle of the ground and walking speed. They concluded that higher Nordic Walking speed increased the force generated by upper limbs on the poles. Furthermore, they suggested changing walking speed rather than the inclination angle of the ground for those patients who are recommended to strengthen upper limbs [12]. Bechard et al. [13], who surveyed 34 patients (12 women and 22 men at the age of 53.6±9.8 years) with degenerative changes of medial compartment of knee joint (after surgical correction of genu varum) using the same measuring device, i.e. poles with force sensors, obtained similar values of axial forces (at the first contact between the pole and the ground – 39.3±24.5 N, and at the stage of pushing the pole against the ground – 14.0±21.0 N).

The comparison of the values of pole-ground force reaction with the values of foot-ground force reaction makes it possible to assess the role of upper limbs in putting a human body into motion while walking with poles. Wilson et al. [10] think that the poles help to increase the walking speed, whereas Schiffer et al. [11] believe that the influence of the force developed by upper limbs on putting the human body into motion seems to be little and that upper limbs are important for keeping the body balance. Shim et al. [14] verified whether there are any differences between Nordic Walking and walking without poles on upper and lower limb muscle activation. Average values and maximum values of the muscle activity of upper limbs were significantly different. However, no differences between Nordic Walking and walking without poles concerning muscle activity of lower limbs were noted.

In order to determine the influence of the poles’ mass on the biomechanical parameters of the walk Schiffer et al. [15] carried out research using poles of different mass. The authors did not identify the influence of the pole’s mass on the values of reaction forces along the long axes of the poles, which amounted from 30.5±16.3 to 71.6±10.5 N, but they showed that the subjective feeling of tiredness increases together with the pole’s mass and that the physiological and biomechanical changes are minor.

Based on the studies carried out to date, it has not been decided explicitly whether the stress exerted on lower limbs, especially on knee joints, while walking with poles is lower than while walking without poles. Hagen et al. [4] used the joined force plates, electrogoniometers and photocells to register the biomechanical quantities while walking with poles. The maximal value of ground reaction force (vertical component) while walking with poles, registered among Nordic Walking instructors, amounted on average to 1.77 of body weight and was similar to the force while walking without poles. Similar conclusions were formulated by Bechard et al. [13] on the basis of the research carried out among patients with degenerative changes of medial compartment of knee joint, although the values of ground reaction force (vertical component) while walking with poles were significantly lower and amounted on average to 0.99 of body weight. In the research carried out by Hansen et al. [16], in which they used a videographic method (APAS system), supplemented with the registration of reaction forces on force plates and the measurement of walking velocity (photocells) to analyse movements while walking with poles, it was concluded that the stress exerted on knee joints during both forms of walking is similar. Based on complex measurements (three-dimensional analysis of walking with the use of a video system integrated with the registration of the ground reaction force and the pole-ground reaction force) carried out among patients with degenerative changes of medial compartment of knee joint, Bechard et al. [13] stated that the moment of force during adducting the limb and the stress put on the knee joint are similar while walking with and without poles. Based on the results of the video analysis and the results from the force plate, Stief et al. [17] showed that the stress exerted on the knee joint, especially in the transverse plane, while walking with poles is bigger than while walking without poles and they concluded that this form of activity should not be recommended when a decrease in the stress exerted on lower limbs is desired.

Variables characterizing the effort made by the disabled in order to propel the wheelchair were evaluated by many authors [18–23] in three main research areas: the mechanics of propelling the wheelchair, the involvement of the locomotor system of the person using the wheelchair (including the risk of overloading upper limbs) and the facilitation of the human effort through the wheelchair construction improvements [24].

The parameters most frequently measured included those determining the developed forces and moments of force, performed work, energy expenditure, power and mechanical fitness, maximal oxygen intake or heart rate [25]. According to Vegter et al. [26], measurements of forces and torques on the handrim are important to study the status and change of propulsion technique with regard to processes of learning, training or wheelchair configuration. The authors [26] compared the outcomes of two different measurementwheels attached to different sides of the wheelchair, determined measurement consistency within and between these wheels given the expected inter- and intra-limb variability as a consequence of motor control. The measured torque around the wheel axle of the two measurement-wheels had a high average cross-correlation of 0.98. Unilateral mean power output over a minute was found to have an intra-class correlation of 0.89 between the wheels. Although the difference over the pushes between left and right power output had a high variability, the mean difference between the measurement-wheels was low at 0.03 W [26].

Crespo-Ruiz et al. [18] carried out a three-dimensional analysis of the kinematics of upper limbs of disabled athletes while propelling the wheelchair during a basketball game using a videographic method. Based on the research, the authors pointed to the need of including the biomechanical parameters in the functional classification of the disabled athletes. Kwarciak et al. [19] suggested a new definition of the cycle of propelling the wheelchair in which – analyzing subsequent periods of contact of the hand with the rims – they considered propelling and resting stages. The average moments of force and power values while propelling the wheelchair by individuals with paraplegia (44 men and 10 women at the age between 18 and 65 years), measured on the axes of the back wheels with the measuring system called SmartWheels, were increasing together with the wheelchair speed (from 1.08 to 1.74 m/s) and amounted, respectively, from 8.98±2.75 to 10.79±3.32 N m and from 32.84±16.96 to 61.84±19.90 W.

The authors [19] believed that further research on the wheelchair propelling cycle should be aimed at identifying potential resting periods and, as a result, dividing the resting stage into separate sub-periods.

The aim of the research carried out by Ambrosio et al. [20] was to determine the influence of the force of the human shoulder joint muscles on the force developed while propelling the wheelchair. Based on the measurement carried out in isokinetic conditions (Biodex system, 5 repetitions of bending-extending the limb in the shoulder joint with angular velocity of 1.05 rad/s (600/s) as well as on the kinematic analysis (Optotrack system) and on the kinetic analysis (SmartWheels system), the authors showed a positive correlation between the force of the human shoulder joint muscles and the force developed while propelling a wheelchair, but they suggested that the training aimed at increasing the muscle force should be separated from improving the technique of propelling a wheelchair.

Ardigo et al. [27] noted the changes in the mechanics of propelling a wheelchair and in the energy expenditure of the disabled basketball players which have occurred together with the technological evolution of wheelchairs since 1960. The authors demonstrated that the improvement in a wheelchair construction lowered both the energy expenditure and the values of mechanical work developed while propelling a wheelchair. It was surprising that the level of fitness of disabled basketball players while propelling modern wheelchairs decreased, which, according to the authors [27], proves that the technological development of wheelchairs related mostly to other factors than energy expenditure, including improvement in stability, maneuverability or grip of the wheelchair.

Chow et al. [21] measured the influence of resistance on the biomechanical characteristic of the wheelchair propelling cycle and on the muscle activity, using a three-dimensional videographic analysis and an electromyography method (EMG). The measurements carried out on the special wheelchair ergometer were used to measure the influence of the inclination angle of the back wheels of the wheelchair towards the ground on the values of the developed force, power and duration, in relation to both the entire cycle of propelling the wheelchair and the contact of the hand with the rim [22]. The research included disabled basketball players who performed three 8-second sprints using a wheelchair with different inclination angles of the back wheels, i.e. 0.16, 0.21, 0.26 rad (9, 12 and 150). After the analysis of the average values of speed (4.28±0.41 m/s), power (81.67±16.76 W), cycle duration (0.41±0.03 s) and propelling stage duration (0.15±0.02 s), it was not determined whether the inclination angle of the back wheels had a significant influence on the mentioned variable. De Groot et al. [28] used a special wheelchair ergometer to determine the influence of the visual feedback on the efficiency of the wheelchair propelling stage among 20 young and healthy men, divided into two equal groups (experimental and control groups). The subjects from the experimental group, who were subjected to the visualization of the value of the developed force in the wheelchair propelling stage, achieved higher values of the mechanical efficiency.

The measurements of the influence of the movement complexity (propelling the wheelchair under different conditions: wheelchair ergometer, conventional treadmill, circular wheelchair track) on the mechanical efficiency and the technique of propelling the wheelchair by healthy individuals did not show any significant correlation between these variables [29]. Wei et al. [23] measured the influence of the wheelchair seat height on the efficiency of the wheelchair propelling cycle and on the kinematics of upper limbs. Based on the analysis carried out with the use of the results of anthropometric measurements, wheelchair ergometer, electrogoniometric measurements and the EMG method, the authors stated that the duration of contact between the hands and the wheelrims increased together with lowering the seat height, and that the bioelectrical activity of forearm muscles was similar. Gagnon et al. [30] quantified the effects of five distinct slopes on spatiotemporal and pushrim kinetic measures during manual wheelchair propulsion on a treadmill at different slopes in individuals with spinal cord injury. As the slope increased, the mean total force was 93% to 201% higher and the mean tangential component of force was 96% to 176% higher than propulsion with no slope. Lui et al. [31] evaluated the mechanical efficiency of two commercially available lever propulsion mechanisms for wheelchairs and compared the mechanical efficiency of lever propulsion with hand rim propulsion within the same wheelchair. Of the two mechanisms, one contained a torsion spring while the other used a roller clutch design. Ten healthy male participants performed submaximal exercise tests using both lever-propulsion mechanisms and hand rim propulsion on two different wheelchairs. Total external power (Pext) was measured using a drag test protocol. Mechanical efficiency was determined by the ratio of Pext to energy expenditure. Results indicated no significant effect of lever-propulsion mechanism for all physiological measures tested. Moreover, results indicated that both lever-propulsion mechanisms tested were more mechanically efficient than conventional hand rim propulsion, especially when slopes were encountered.

Other authors [32] compared the force application characteristics at various push frequencies (at 60, 80, 120 and 140% of freely chosen frequency) of asynchronous and synchronous hand-rim propulsion. Both asynchronous and synchronous hand-rim propulsion demonstrated similar trends: changes in push frequency are accompanied by changes in absolute force even without changes in the gross pattern/trend of force application and gross efficiency.

While analyzing scientific publications in which forces developed in the wheelchair propelling cycle were measured, Kwarciak et al. [19] pointed to the limitations encountered during the measurement. They listed such most important limitations: measurements carried out under artificial conditions, other than random choice of subjects, abnormal division of measuring data, not having excluded the influence on the registered measurement results of many variables, e.g. level of sensory abilities and muscle strength of the subjects or a wheelchair construction. In a review article, Dellabiancia et al. [33] showed that the combined and contemporary use of the entire instrumentation group is the best strategy, providing a great amount of data to achieve an easier and accurate analysis of wheelchair propulsion. The authors confirmed that the procedure most commonly used for the analysis of articular forces and moments involved the use of SmartWheel and also specified the model “OptiPush” [33]. In order to optimize both the technique of propelling the wheelchair and the wheelchair construction, further research taking into account biomechanical and physiological aspects is needed [22].

The usefulness of measuring forces exerted by a physiotherapist on the patient’s body while applying manual mobilization techniques was mentioned in many studies [34, 35, 36, 37]. It was underlined that the values of forces intuitively applied by physiotherapists might not only be inappropriate, but also dangerous for the patient [38]. To date, no method of direct measurement of forces exerted by a physiotherapist on the patient’s body has been drawn up, and that is why indirect methods are applied, which make it possible to register the reaction forces with the use of a special measuring table or mannequin [39–44], spine simulator [45] or a mathematical analysis employing different models of muscles [38]. Other authors noted the effects of manipulation on spine neuromechanical responses by EMG [46].

After the analysis of the literature relating to manual mobilization techniques used in physiotherapy, Snodgrass et al. [34] stated that an intuitive, subjective proportioning of force exerted by a physiotherapist upon the patient’s body was a serious limitation, when it comes to the standardization of both mobilization techniques and methods of measuring force, which can negatively influence the efficiency of the treatment. Following this analysis, the aforementioned authors [34] organized a measuring station to register and visualize – in the on-line system – the values of forces developed by a physiotherapist while applying manual mobilization techniques [35, 36]. The aforementioned station designed as a measuring table with seven two-axis force sensors made it possible to carry out a three-dimensional measurement of forces exerted by the physiotherapist upon the patient’s body [36]. Snodgrass et al. [35], who made measurements with the use of the station, determined average values of the maximal force developed by a physiotherapist while applying the back-front (central and unilateral) level I– IV mobilization on cervical vertebrae (C2–C7), which amounted to 21.8±15.0, 34.9±20.9, 58.2±27.5 and 61.0±29.9 N, respectively. Significantly higher values were registered by Chester et al. [46] while exerting pressure on the third lumbar vertebra, which amounted to 350 N with the use of level IV mobilization with the Maitland method. These values should be treated as approximate ones, because they are registered – as underlined by the authors – under artificial conditions, i.e. towards an inflexible surface of the measuring device, rather than towards the patient’s body [36]. In order to measure and monitor the forces developed by a physiotherapist while applying manual mobilization techniques, Waddington et al. [37] and Waddington and Adams [47] used a specially designed prototypical manual dynamometer. During the back-front manual spine therapy, the values of force developed by 30 physiotherapists fit in the range between 50 (level I) and 180 N (level IV). An attempt was made to measure indirectly the force exerted by the physiotherapist upon the patient’s body while applying manual mobilization of the lumbar and cervical spine [39, 41, 43]. Individuals applying this mobilization on a special measuring table used the feedback and continuously monitored average values of force and frequency of its development on the screen. Sheaves et al. [40] proved the significance of a continuous control of the force developed by the physiotherapist in learning a manual technique of mobilization of the lumbar spine. In the research encompassing 116 experienced physiotherapists and 120 physiotherapy students [41] in which a special measuring table was used to enable a continuous observation of the force developed during manual manipulation, it was proved that experienced physiotherapists developed significantly higher forces than students. It was also concluded [41] that the forces developed by physiotherapists depended on the patients’ sex and body weight, because bigger forces were exerted upon men and individuals with higher body weight.

When describing the forces developed by chiropractors performing spinal manipulation, Herzog [48] pointed to considerable differences in the values of forces between individuals using this treatment.

While evaluating changes in force-time characteristics developed by physiotherapists performing spinal manipulation in the period of 5 years of teaching, Descarreaux and Dugas [42] used a measuring station including a mannequin and a force platform, which physiotherapists were standing on when applying the treatment. In the teaching period under examination the authors revealed a different pace of changing the parameters describing the applied manipulation techniques.

Myers et al. [49] suggested a new method of predicting spinal contact force which combines direct (load cell, motion capture) and indirect (force plate) measurements into a single framework and preserves clinically relevant practitioner-participant contacts.

The aim of the research carried out by De Souza et al. [44] was to identify the correlation between the developed force and the change of the patient’s body position during the front-back mobilization of the ankle joint. To register the forces developed by the physiotherapist the authors used a miniaturized force platform. Proving the usefulness of applying the feedback during manual mobilization (the values of force were observed on-line on the screen by the physiotherapist), they underlined that such a measure contributes to the minimization of differences in proportioning forces between individual physiotherapists while applying the same techniques, which translated into better accuracy of the treatment.

Stemper et al. [50] used a measuring station equipped with an electro-hydraulic mannequin to measure the values of the pressure force during manual therapy applied to the thoracic spine. Although the values of the pressure force exceeded 1000 N in individual cases, the authors stated that these were values amounting to ca. 20 % of the acceptable limit of the pressure exerted on the chest.

The physiotherapist applying a traction technique in hip and shoulder joints developed relatively high values of force [51, 52]. Vaarbakken and Ljunggren compared the efficiency of the traction performed in the hip joint with different values of forces used to relieve the pain resulting from the degenerative changes in the joint. Every day for one month the physiotherapists learned how to feel the force developed during the traction while working with a model of a lower limb with a calibrated value of the developed force, and having achieved the accuracy amounting to 50 N they made measurements on patients. It was proved [51] that a gradual increase in force up to 800 N (acting with the so-called full force) brought a better therapeutical effect than the traction with lower (standard) values of force.

Based on the radiological research, Gokeler et al. demonstrated that the traction applied in a shoulder joint, using 137 N for 40 s, did not change the distance between the humeral head and the surface of the glenoid cavity of the scapula.