Musculoskeletal and finite element modelling are often used to predict joint loading and bone strength within the human hand, but there is a lack of in vitro evidence of the force and strain experienced by hand bones.
This study presents a novel experimental setup that allows the positioning of a cadaveric digit in a variety of postures with the measurement of force and strain experienced by the third metacarpal. The setup allows for the measurement of fingertip force as well. We tested this experimental setup using three cadaveric human third digits in which the flexor tendons were loaded in two tendon pathways: (1) parallel to the metacarpal bone shaft, with bowstringing; (2) a semi-physiological condition in which the tendons were positioned closer to the bone shaft.
There is substantial variation in metacarpal net force, metacarpal strain and fingertip force between the two tendon pathways. The net force acting on the metacarpal bone is oriented palmarly in the parallel tendon condition, causing tension along the dorsum of the metacarpal shaft, while the force increases and is oriented dorsally in the semi-physiological condition, causing compression of the dorsal metacarpal shaft. Fingertip force is also greater in the semi-physiological condition, implying a more efficient grip function. Inter-individual variation is observed in the radioulnar orientation of the force experienced by the metacarpal bone, the fingertip force, and the strain patterns on the metacarpal shaft.
This study demonstrates a new method for measuring force and strain experienced by the metacarpal, and fingertip force in cadaveric digits that can, in turn, inform computation models. Inter-individual variation in loads experienced by the third digit suggest that there are differences in joint contact and/or internal bone structure across individuals that are important to consider in clinical and evolutionary contexts.
Different manipulative and locomotor behaviours lead to varied biomechanical loading of the hand and affect the external morphology (e.g. entheses) and internal structure (e.g. cortical thickness, trabecular architecture) of bones. The ability of bone tissue to remodel in response to the magnitude and direction of load is known traditionally as Wolff’s law (
With recent advances in 3D imaging techniques, the internal bone structure of human, nonhuman primate, and fossil human (hominin) hand bones has been investigated to help reconstruct hand function and behaviour in extinct taxa. Variation in cortical and trabecular bone structure has been linked to the differences in hand loading during locomotor behaviours in living apes vs. manipulation in humans (
Understanding the loads experienced by the hands is also important within a clinical context, including research on joint disease (
Because both in vivo and in vitro studies of hand bone loading present several technical challenges, musculoskeletal models are typically used to predict force during isometric hand functions, such as pinch and power grip tasks (
The purpose of this study is to help to fill this gap of information on in vitro hand bone loading. Here, we present a novel experimental setup that allows simultaneous measurement of in vitro loading and deformation of human and nonhuman primate hand bones during functional postures. The experimental setup also allows the measurement of fingertip force. This experimental design builds upon previous in vitro studies that typically only measure fingertip force in cadaveric specimens (
Two clamps were designed to fixate a cadaveric digit at its fingertip and its metacarpal base. The concave cavities of the clamps were designed to hold the fingertip and the metacarpal base of the specimen (see the distal and proximal clamps in
A customized experimental setup was designed to fix the fingertip and the base of metacarpal bone using the distal and proximal clamps, respectively. A six-axis load cell was fixed onto the proximal clamp and the proximal stand to measure the force experienced by the metacarpal bone, and a set of pulleys and weights were used to load the tendons to simulate the muscle contraction during hand function. Photo credit: Dr. Szu-Ching Lu.
(A) The load cell was attached to the proximal clamp to measure the force experienced by the metacarpal bone, and the strain gauges were attached to the metacarpal shaft to measure the bone deformation. The third digit was in a flexed posture with the flexor tendons loaded and guided parallel to the bone shaft, and then a low-friction metal bar was applied to place the tendons in a semi-physiological pathway (the dash line). (B) The load cell was fixed to the distal clamp for the fingertip force measurement. DIP, distal interphalangeal joint; PIP, proximal interphalangeal joint; MCP, metacarpophalangeal joint. Photo credit: Dr. Szu-Ching Lu.
The experimental setup also included eight pulleys for loading tendons to simulate muscle contraction during hand function (
Only one load cell was used in this study and therefore the net force experienced by the metacarpal bone and the fingertip force had to be measured separately. To measure the metacarpal force, the load cell was attached to the proximal clamp and the proximal metal stand. During the metacarpal force measurement, the bone deformation was quantified at the same time using strain gauges (FLA-1-11-1L; Tokyo Sokki Kenkyujo Co., Ltd., Tokyo, Japan). A compact data acquisition system (NI cDAQ-9174; National Instruments, Austin, TX, USA) and a customized program (LabVIEW; National Instruments, Austin, TX, USA) were designed to acquire the metacarpal force and strain signals simultaneously. To measure the fingertip force, the load cell was attached to the distal clamp and the distal metal stand. The same data acquisition system was used to acquire the fingertip force data.
Although the experimental setup can be used for each digit, we tested the setup on the third ray as this digit plays an important role in the daily activities of humans (
(A) Three strain gauges were applied to the radial-palmar, dorsal and ulnar-palmar sides of the metacarpal bone at its midshaft to quantify the bone deformation. (B) A computed tomography image shows the coronal cross-sectional view of the metacarpal bone with three strain gauges attached. Photo credit: Dr. Szu-Ching Lu.
With the fingertip and the base of metacarpal bone fixed, each specimen was positioned with its distal interphalangeal (DIP) joint at 25° flexion, the proximal interphalangeal (PIP) joint at 60° flexion, and the MCP joint at 55° flexion as this joint configuration is at the approximate midpoint of the functional range of motion required to perform 90% of daily manipulative activities (
Unlike previous studies that only measured the fingertip force in cadaveric specimens (
Following the quantification of the force and strain experienced by the metacarpal, the strain gauge wires were removed and the load cell was fixed to the distal clamp and the distal stand for the direct measurement of fingertip force. The same posture, loading conditions and tendon pathways were applied as described above.
First, the force and strain experienced by the metacarpal bone were collected synchronously at 100 Hz, and then just fingertip force data was acquired at 100 Hz. The acquired force and strain signals were calibrated with baseline data collected when the flexor tendons were not loaded. Then, the signals were filtered using a sixth-order Butterworth low-pass filter with the cut-off frequency at 6 Hz. For the parallel condition, the average force and strain values within the middle 3 seconds of data collection were calculated. For the semi-physiological condition, a single frame of data was extracted since the tendon path was dynamically modified. All analyses were conducted within a customized MATLAB script (The MathWorks, Inc., Natick, MA, USA; see
The force and strain results varied substantially between the two tendon path conditions and, in some cases, across individuals. Below we present the results of net force and strain experienced by the metacarpal bone and fingertip force of the three test subjects in each tendon path condition.
The net force acting on the third metacarpal increased when the tendon path was changed from parallel to the semi-physiological condition. The resultant force increased from 3.93 ± 0.23 N in the parallel condition to 5.50 ± 0.47 N in the semi-physiological condition (
Tendon pathway | Proximal (+) distal (−) | Dorsal (+) palmar (−) | Radial (+) ulnar (−) | Resultant |
---|---|---|---|---|
Parallel | 3.86 ± 0.26 N | −0.58 ± 0.07 N | 0.34 ± 0.37 N | 3.93 ± 0.23 N |
Semi-physiological | 5.29 ± 0.40 N | 1.23 ± 0.57 N | 0.42 ± 0.74 N | 5.50 ± 0.47 N |
The force experienced by the third metacarpal bone is presented in the dorsal-palmar, proximal-distal and radial-ulnar directions with respect to the metacarpal bone, and the resultant force magnitude is also presented. The force values are presented as mean ± standard deviation in Newtons.
The force is depicted in the flexion-extension plane (A) and the radial-ulnar deviation plane (B) with respect to the third metacarpal bone. The arrow shows the mean value of the three specimens and the box shows the range of one standard deviation. Data of each specimen are also presented.
The strain experienced by the dorsum of the metacarpal bone was consistent with the direction of the net force experienced by the metacarpal. In the parallel tendon condition, the dorsal side was in tension as the force was oriented palmarly, while in the semi-physiological condition, the dorsal side was in compression as the force was oriented dorsally (
The strain experienced by the three sides of the metacarpal bone is presented as the mean value (box) with one standard deviation (whiskers). Data of individual specimen are also presented. Positive value represents tension while negative value means compression.
Similar to the pattern found for metacarpal force, fingertip force also increased as the tendon path changed. The mean value of the fingertip force across the three specimens increased from 1.02 ± 0.10 N in the parallel condition to 1.67 ± 0.06 N in the semi-physiological condition (
Tendon pathway | Proximal (+) distal (−) | Dorsal (+) palmar (−) | Radial (+) ulnar (−) | Resultant |
---|---|---|---|---|
Parallel | −0.39 ± 0.10 N | −0.82 ± 0.07 N | −0.22 ± 0.52 N | 1.02 ± 0.10 N |
Semi-physiological | −0.33 ± 0.14 N | −1.51 ± 0.10 N | −0.26 ± 0.66 N | 1.67 ± 0.06 N |
The fingertip force is presented in the dorsal-palmar, proximal-distal and radial-ulnar directions with respect to the distal phalanx. The resultant force magnitude is also presented, and the values are presented as mean ± standard deviation in Newtons.
The force is presented in the flexion-extension plane (A) and the radial-ulnar deviation plane (B) with respect to the distal phalanx. The arrow shows the mean value of the three specimens and the box shows the range of one standard deviation, and the data of individual specimen are also presented.
We present a novel experimental design that is able to simultaneously measure in vitro force and strain experienced by the metacarpal bone without disrupting the MCP joint space. In addition, this experimental system is able to measure the fingertip force. This experimental approach builds upon previous in vitro studies have typically measured fingertip force only (
We found substantial variation the direction of net joint force and metacarpal shaft strain between the parallel and semi-physiological pathway conditions. Although force primarily loaded the metacarpal proximally in both tendon conditions, force was lower and oriented more palmarly in the parallel condition compared to higher force oriented dorsally in the semi-physiological condition. During the experiment, care was taken to avoid touching the metacarpal shaft when the metal bar was applied to change the tendon path from parallel to semi-physiological condition. However, we observed that the flexor tendons wrapped around the metacarpal head and pushed the bone dorsally, which may lead to the higher net force acting dorsally on the metacarpal bone in the semi-physiological condition. Accordingly, the strain experienced by the dorsum of the metacarpal changed from tension in the parallel condition to compression in the semi-physiological condition. In addition to the net force acting on the metacarpal bone, the fingertip force also increased in the semi-physiological condition, and the fingertip force was more proximally-oriented. The change in fingertip force magnitude and orientation may also be a result of the flexor tendons wrapping around the metacarpal head and applying additional force dorsally. The fingertip force increased even though the moment arm of the flexor tendons with respect to the MCP joint rotation centre decreased, suggesting a more efficient grasp function when the tendons run closer to the metacarpal bone shaft. This difference in force and strain between the two tendon pathway conditions also reveals the importance of tendon path simulation in computational models, which are typically modelled in a simplified manner similar to parallel condition (
Variation in radial-ulnar force experienced by the metacarpal bone was observed in this study. There may be several reasons for this variation. Force in the radial-ulnar direction accounted for only 2–16% of the primary proximally-oriented force. The orientation of force in the radial-ulnar direction could be influenced by subtle variation in the asymmetry of the external shape of the metacarpal head (
Our results on fingertip force are generally consistent with those of the few previous studies that have investigated in vitro fingertip force in a human finger (
This study has several limitations, with the small sample size and the in vitro testing condition as the most important factors. However, we consider three individuals to be sufficient to demonstrate the value of the newly developed experimental setup described here, which is able to simultaneously measure in vitro metacarpal force and strain. While the results of in vitro measurement might deviate from the in vivo condition, in vivo assessment of loading and deformation of hand bones is inaccessible with current technology and ethical regulations. In addition, this study not only quantified the force and strain experienced by the metacarpal bone but also measured the fingertip force, providing a more comprehensive data set for comparing the in vitro measurement of this study to the in vivo fingertip force within the literature (
Another potential limitation is the relatively low tendon loads (3% of the maximal muscle force) that were applied to the specimens to avoid rupture of the tendon, thus resulting in small force and strain measurements. Accuracy error of the sensors and the noise in the signals may bias the results. However, the accuracy error of the load cell was less than 0.05 N (see
This study used the same tendon load for three specimens while there might be individual difference in the anthropometry of the soft tissues. Due to the logistical challenges of measuring the PCSA of each specimen prior to the experiment, this study used the published mean PCSA values of the human FDS and FDP (
Finally, the parallel and semi-physiological tendon pathways used in this study may not be ideal for simulating the tendon path within a hand with intact soft tissues. The parallel tendon pathway is a simplified condition that, although not fully representative of the biological condition, is particularly valuable for building informative musculoskeletal or FE models and obtaining replicable predictions. Although the artificial positioning of the tendons and the data extraction method could be improved, this study provides the best possible approximation of the normal physiological state within the current experimental constraints.
Although this study presents the force and strain measured in human third digit only, the experimental setup is adjustable to fit variation in the size and morphology of other human and nonhuman primate digits, as well as varying digit postures. Future studies could also include loading of more tendons and/or different loading patterns. Comparison among different digits with varied functional postures and tendon loading patterns may inform research on osteoarthritis, prosthetic design, and the interpretation of internal bone structure within the human hand. Comparative studies between humans and nonhuman primate digits can help inform the interpretation of the differences in external and internal bone morphology in relation to locomotor and manipulative behaviours. This comparative context is needed for more robust functional interpretations of fossil hominin hand morphology and the reconstruction of hand use throughout human evolution.
This study presents a novel experimental design for simultaneous in vitro measurement of metacarpal force and strain. In addition, the experimental design allows for the quantification of fingertip force. Testing this experimental setup on three human cadaveric fingers revealed substantial difference in force and strain between two tendon pathway conditions, which, in turn, emphasizes the importance of tendon path simulation in computational models. The in vitro data presented in this study are not only useful for musculoskeletal and FE modelling in hand-related research such as prosthesis design, but also for the functional interpretation of the variation in internal bone structure within humans. In the future, investigation of forces and strain in different fingers could contribute to a better understanding of the varied prevalence of the MCP joint osteoarthritis across the digits (
This table presents the force and strain data measured from the three specimens in two tendon path conditions, i.e. parallel and semi-physiological conditions. The force experienced by the third metacarpal bone (Fmc) is presented in the proximal(+)/distal(−), dorsal(+)/volar(−), and radial(+)/ulnar(−) directions with respect to the metacarpal bone. The strain gauge (SG) measurement was performed at three sides of the bone, i.e. radial-palmar, dorsal, and ulnar-palmar sides. The fingertip force (Ftip) is presented in the proximal(+)/distal(−), dorsal(+)/volar(−), and radial(+)/ulnar(−) directions with respect to the distal phalanx.
The load cell was tested before being used in the experiment. The error was less than 0.05 N and the full-scale accuracy error was less than 0.1%.
The precision of strain gauge measurement was verified via beam models, and the analytical calculations showed the same strain pattern and similar values as the experimental results. Photo credit: Dr Szu-Ching Lu.
The acquired force and strain signals were calibrated with baseline data collected when the flexor tendons were not loaded. Then, the signals were filtered using a sixth-order Butterworth low-pass filter with the cut-off frequency at 6 Hz.
We thank D. O’Connell, A. Brookman and H. Twyman from the Engineering Workshop, School of Engineering and Digital Arts, University of Kent for their help in developing the experimental system, and we thank C. Dunmore for his help in processing the computed tomography images. We also thank the medical students who helped prepare the cadaveric specimens at the Jan Palfijn Anatomy Lab (University of Leuven). We are grateful to the Editor and two anonymous reviewers for their thoughtful comments that improved this manuscript.
The authors declare that they have no competing interests.
The following information was supplied relating to ethical approvals (i.e. approving body and any reference numbers):
The human cadaveric hands were obtained through the Human Body Donation Programme from the Medical Faculty of the University of Leuven, Belgium.
The following information was supplied regarding data availability:
The raw data are provided in the