The effect of grip force,stroke rotation and frequency on discomfort for a torqueing tasks

Abstract

BACKGROUND:

Occupational tasks involve awkward upper limb postures, especially movement of forearm with repetitive combined gripping and torqueing exertions, which may lead to development of WMSDs. From the literature survey it was observed that there was a lack of studies focussed on the combined effect of torque and grip exertions on forearm discomfort.

OBJECTIVE:

The present study was to investigate the effects of grip force, stroke rotation and frequency of exertions on discomfort and Electromyography (EMG) activities of the forearm muscles in a repetitive torqueing task.

METHOD:

Twenty-seven male participants volunteered in this study. The participants performed repetitive exertions for a 5 minutes duration for each combination of the different levels of stroke rotation, grip force and frequency of exertions. Three levels of stroke rotation, three levels of grip force and three levels of frequency of exertion were chosen as independent variables. Therefore a 3 × 3 customized factorial design was used for the experiment for each level of grip force. Hence, the study was divided into three groups on the basis of grip force (50N, 70N and 90N).

RESULTS:

The ANOVA showed that stroke rotation and frequency of exertion were significant on discomfort. Further Students Newmann test (SNK) revealed that discomfort was increased with increasing stroke rotation and frequency of exertion. The multivariate analysis of variances (MANOVA) performed on EMG data instead of ANOVA because EMG activities of five muscles simultaneously were recorded. The Results found that extensor muscles were more fatigued in torqueing with gripping task.

CONCLUSIONS:

It was found that stroke rotation for the torqueing tasks must be kept below 45°. It was concluded that it is important to control stroke rotation to improve performance of repetitive torqueing activity.

Keywords

Discomfort grip force muscle fatigue stroke rotation repetitive torqueing exertion

1 Introduction

Several studies have been conducted on the prevalence of upper extremity disorders in repetitive hand-intensive work [1 –3]; due to awkward working postures [4 –6]; repetitive exertions with higher force levels of torque, pinch grip etc. [7, 8]. Due to prevalence of upper extremity disorders in industries many researchers conducted experiments related to tasks of various types of industries for example Hughes et al. [9] studied on 104 workers at an aluminium smelter, they found a prevalence rate of 11.6% for elbow and forearm injuries. In the workers of garment industry Herbert et al. [10] reported 29% prevalence of pain symptoms in forearm and elbow. The human powered hand tools used in material handling and force/torque exertions in all industries of United States were responsible for approximately 45% of injuries due to overexertion [8, 11].It is also reflected by the expenditures of United States for work-related musculoskeletal disorders (WMSDs) costing approximately $50 billion per year [12].

From the survey, it was noticed that the factorsaffecting discomfort, productivity and risk of WMSDS were force, posture and repetitions [5 , 14]. Therefore many researchers conducted experiments to evaluate the effect of these factors on discomfort in industrial tasks. For example, Li and Yu [15] reported that there was a strong correlation with grip force and discomfort for different postures in gripping task. Also, Lin and McGorry [16] developed models for grip force which was associated with discomfort and found that grip force was significant on discomfort. In an assembly task, O’Sullivan and Clancy [17]; Finneran and O’Sullivan [18] reported that force and repetition was highly significant factors on discomfort. Also, in torqueing task, Mukhopadhyay et al. [19] found that posture, torque and frequency were significant on discomfort. In a simulated drilling task, Mehta and Agnew [20] very nicely explained the difference of gender and age in task performance. They have given a good outcome of distribution of fatigue/discomfort over hand arm system including shoulder. Although they have not considered any angular deviation that is very likely to have possible contribution in developing discomfort for a task spread over a wall. That needs to be further evaluated.

In line of literature survey it was reported that forearm rotation plays an important role in industries by which discomfort and risk of WMSDs were affected. O’Sullivan and Gallwey [21] reported that the more common elbow and forearm injuries, including humeral epicondylitis and pronator teres syndrome (both medial and lateral) have been linked to forearm rotations. They also found that forearm rotation significantly affected MVC torques during repetitive exertions. In a study of engineering plant, Dimberg [22] found 7.4% workers have epicondylitis syndrome which was directly associated with the task involving forceful exertions. In another laboratory experiment based on intermittent isometric torque exertions, O’Sullivan and Gallwey [23] reported both forearm angle and twisting direction were significant on forearm discomfort (p < 0.001). Also in torqueing task [24], flexion task [25 –27] and gripping task [28] found a significant effect of forearm rotation on discomfort.

The literature survey showed that in industrial tasks, torque has a strong association with musculoskeletal disorder and injuries. Therefore many researchers examined the effect of torque on hand-arm stress in an in-line screw driver [29 –31] and some had conducted experiments in which they evaluated the effect of torque on forearm discomfort [8 , 24]. It was found from the above studies that pronation/supination torques significantly increased discomfort.

Many Industrial tasks involve awkward upper limb postures, generally hand tools, in which movement of forearm with repetitive combined gripping and torqueing exertions involved. From the literature survey it was observed that there was lack of studies focussed on the combined effect of torque and grip exertions on forearm discomfort. However, only Mukhopadhyay et al. [32] investigated the combined effect of grip force and intermittent isometric torque on forearm discomfort at different combinations of shoulder abduction (0° and 90°) and forearm rotation angle (0±60% ROM). They conducted an experiment at an isometric forearm torque (20% MVC supination) and applied grip force (0 or 20% MVC) at different combinations of shoulder abduction angle and forearm rotation angle. They found that the combined torque and gripping exertions were highly significant on discomfort (p < 0.001). Therefore, it was necessary to prevalent the upper limb related WMSDs in combined repetitive gripping and torqueing task for different levels of frequency of exertions which was frequently used in assembly industries. The objective of the present study was to investigate the combined effect of grip and torque in repetitive upper limb exertions for three levels of grip force, three levels of stroke rotation and three levels of frequency of exertion on forearm discomfort. Therefore the Null hypothesis for the present study was as follows:

“ there was no main and/ or interaction effects of the levels of grip force, stroke rotation and frequency of exertions on discomfort and EMG activities of extensor and flexor muscles of forearm”.

2 Methodology

2.1 Participants

Twenty seven right handed university male college students participated in this study. Their mean age μ= 24.48 years (σ= 3.40), weight μ= 63.5 kg (σ= 8.99) and height μ= 171.35 cm (σ= 6.60). All participants had reported no history of musculoskeletal/another type of injuries in hand arm system in the past. They were called by notices in the college to volunteer in this experimental investigation.

2.2 Experimental design

In the experiment participants performed repetitive upper limp exertions at three levels of stroke rotation (30°, 45° and 60°), three levels of grip force (50N, 70N & 90N) and three levels of frequency of exertion (10 exertions/minute, 15 exertions/minute and 20 exertions/minute). A 3 × 3 customized factorial design with 27 combinations was used. The study was divided into three groups (each group having 9 participants) on the basis of grip force; 50N, 70N and 90N, and each group performed the repetitive exertions in 5 mins at each level of frequency and stroke rotation (i.e., 9 treatments for each participant of each group). after each experimental treatment a gap of at least five minutes or till participant had no discomfort was given before starting the next treatment (as in the line of [23]). The independent variables were taken to investigate the effect of these on EMG activities of muscles and discomfort. Experiment was conducted in a unique random order for every participant.

2.2.1 Independent variables

Stroke rotation. The three levels of forearm rotations angles (60% prone and supine, and neutral range of motion) for intermittent isometric pronation torque [8, 19] and for MVC pronation torque [24]. O’Sullivan and Gallwey [21] conducted experiment on 75% prone ROM, neutral and 75% supine ROM forearm rotation for maximum supination and pronation torques. O’Sullivan and Gallwey [23] considered 11 forearm rotation angles at the increment of 15% ROM from 75% prone to 75% supine i.e 75%, 60%, 45%, 30%, 15% pronation range of motion, neutral and 75%, 60%, 45%, 30%, 15% supination range of motion for exerting isometric torque. In the previous in-house experiment [33] participant performed repetitive screw driving task for 2 minute, stroke rotation were taken as 30°, 60° and 90° in supination direction. Because of the inability of doing the task at stroke rotation 90° as discomfort was found highest at 90° stroke rotation therefore in the present study three levels of stroke rotation 30°, 45° and 60° from neutral position in the supination direction were used.

Grip force. With reference to previous studies [15 , 34–36] and keeping in mind the requirements of the occupational tasks in general, grip force levels of 50, 70 and 90N were considered for the present study. Instead of relative grip force absolute values were chosen because in most of the industries task demands a specific value of grip force rather than the percentage of individual’s maximum force.

Frequency of exertion. In line of previous studies [8 , 24], in the present study 10, 15 and 20 exertions/minute levels of frequency of exertions were considered for task investigations.

2.2.2 Dependent variables

Electromyography (EMG). Hoozemans and Dieën [35] examined flexor carpi radialis (FCR), flexor digitorum superficialis (FDS), extensor carpi ulnaris (ECU), extensor digitorum (EDI), extensor carpi radialis brevis (ECRB) and extensor carpi radialis longus (ECL) muscle’s EMG activities for an isometric gripping task. O’Sullivan and Clancy [17] considered flexor carpi radialis (FCR) and flexor carpi ulnaris (FCU) to determine EMG activity in repetitive assembly task. In a gripping task, Mogk and Keir [36] quantified the effects of wrist, forearm and grip force on forearm muscles. They examined the EMG activities of FCR, FCU, FDS, ECR, ECU and extensor digitorum communis (EDC). Based on the reporting and findings of the above stated studies in the present study, it was decided to record the surface EMG activities of five forearm muscles (FCR, FCU, FDS, ECR, ECU).The electrodes (EMG pre amplifier) were attached to the cleaned surface of forearms on the locations as suggested as per Grey’s Anatomy under the supervision of the University faculty member of the department of Anatomy. The DataLINK apparatus of the Biometrics Ltd. (UK) was used for acquiring surface EMG signals for these five muscles. The EMG signals were recorded at the sampling rate of 1024 Hz using Surface EMG sensor (Model: SX230 EMG sensor; Make: Biometrics Ltd. UK). The pre-amplified signal of EMG was interfaced to the Laptop (HP based on Dual-Core Processor) using 8 channel subject unit of DataLINK (DLK900). The signal was conditioned using the DC & Low frequency filters provided in the same software.

Discomfort. To record the perceived discomfort of the participant for experimental task, a code was written in LABVIEW 8.6 that assist the participant to give discomfort on a 100 mm long Visual Analogue Scale [8 , 37].

2.2.3 Apparatus

Grip meter. A digital grip force Dynamometer (Model: Precision Dynamometer G200; Make: Biometrics Ltd. UK) was interfaced to the DataLINK (Make: Biometrics Ltd. UK). DataLINK was connected with the Laptop through USB1800 connecting lead. The data of grip levels were recorded at the sampling rate of 50 samples/second as recommended by M/s Biometrics Ltd. The levels of out of range efforts were alarmed using different colours and a buzzer in the software settings.

2.2.4 Experimental setup

A wheel with handle was integrated in the rig fixed at a platform as shown in Fig. 1. A spring was attached to the rod of the rig to return back the handle in its starting position. For the angle of rotation of forearm a 360° range protector was fixed on the rig and a pointer was placed to indicate the value of rotation. These details are labelled in Fig. 1.

2.2.5 Task

In the experiment every participant had to exert the given level of grip force for one second duration followed by the clock shown on the screen. Simultaneously the participant was asked to rotate the grip dynamometer in supine direction for the respective stroke rotation as per the treatment combination displayed randomly on the screen. This exertion was to be performed at the frequencies of 10, 15 and 20 exertions/minute. The frequency was controlled using the clock displayed to the participant on the computer screen with three hands, one rotating (green) and other two (red) to mark the start and end of the exertion label (Fig. 2).

2.2.6 Procedure

Participants were briefed about the experiment and signed a prior consent. The experiment was divided into three groups which were based on grip force levels. Hence nine treatments were assigned for each participant for only one of the levels of grip force. After preliminary attachment of sensors and apparatus, the participant sat on the adjustable chair at a fixed position on the floor with regard to the experimental rig (Fig. 1). The elbow flexed at 90° without upper arm adduction and with neutral forearm angle. The maximum and minimum EMG activities were recorded for respective muscle to normalise EMG data. For a respective combination, participant exerted a 50N force using grip dynamometer in supination direction e.g. from 0° to 30° in one stroke for one second duration then released, rested for 5 seconds i.e. 10 exertions/minute, repetitively for 5 minutes duration at each articulation as followed by the clock shown in Fig. 2. The time was controlled by this analogue clock on the computer screen and started the green light at the end of each treatment. At the end of each 5-min treatment the participant’s discomfort was recorded on the VAS displayed on the computer screen and the EMG data stored for further analysis. The participant was told that the symptoms of discomfort could include numbness, warmth, cramping, pulling, soreness, fatigue aching, tenderness, pressing or pain [26, 38]. The screen shot of recording the discomfort with the EMG activities of five forearm muscles and grip force simultaneously in Fig. 2 on left side of the screen. A rest of at least 5 min was provided between the treatments or until the until the participant reported no feeling of discomfort. Total time required for one treatment was 10 minutes therefore total time required for the entire treatments was 80 minutes for one participant. The same procedure was repeated for group 2 (participants recruited in the experiment for grip force 70N only) and group 3 (participants recruited in the experiment for grip force 90N only).

2.3 Statistical analysis

All statistical analyses were performed using SPSS 19. To evaluate the effect of grip force, stroke rotation and frequency of exertion on discomfort, a three way repeated measures analysis of variance (ANOVA) was applied. While to examine the effect of independent variables on EMG activity, multi-variant analysis of variance (MANOVA) was applied. Further Student Newman Keuls post hoc test was applied to explore the significant effects of different levels of main factors on response variables. Before ANOVA/MANOVA for discomfort and EMG signals one way ANOVA was performed considering order number as independent variable and main dependent variables as it is in the present study. In some controversial cases (i.e. differences of points seen on graphs in profile plots) in the profiles plotted for discomfort score and other parameters of EMG signal with respect to independent variables, pairwise t-tests were performed to evaluate significant differences in those points of different combinations of independent variables.

3 Results

3.1 Discomfort

The summary of raw discomfort score (RDS) was presented in Table 1. The range of RDS was found approximately from 2 units to 7 units. The minimum RDS (μ= 2.24; σ = 1.57) for a condition at 90N grip force with 60° stroke rotation at frequency of 10 exertions/minute. However, maximum perceived RDS (μ = 7.07; σ = 2.18) was found at 50N grip force, 60° stroke rotation and frequency of 20 exertions/minute. Increasing trend of RDS for 50 and 70N grip force was found with the increased value of stroke rotation. For grip force 90N the mean value of RDS was found increased when moved from stroke rotation 30° to 45°and then was decreased from 45° to 60° stroke rotation.

The present study involved nine participants for each group, hence the perception of discomfort score of participants were different for each combination. Therefore in line with Carey and Gallwey [39], the data need to be standardised using min-max standardisation procedure [40] as follows:

Standardised discomfort score $\begin{matrix} {(SDS)}_{ij} & = & \frac{(raw {data}_{ij} - \min, {data}_{j})}{(\max . {data}_{j} - \min . {data}_{j})} \\ \frac{(raw {data}_{ij} - \min . {data}_{j})}{(\max {data}_{j} - \min . {data}_{j})} \times 10 \end{matrix}$

The SDS was calculated using above mentioned formula and presented in Table 2 were used for further statistical analysis.

The Table 2 showed the summary of standardised discomfort score (SDS). The range of SDS was found to be around 2.5 units to 9.0 units. The maximum SDS noticed at 90N grip force with stroke rotation 60° and frequency of exertion 15 exertions/minute (μ = 9.0; σ = 1.63). The minimum SDS was at 70N grip force with 30° stroke rotation and 10 exertions/minute (μ = 2.53; σ = 3.84). The mean SDS with respect to frequency 10 exertions/min, 15 exertions/min and 20 exertions/min were found as 4.38, 6.43 and 6.35 units respectively. As well as mean SDS for grip force 50N, 70N and 90N were found to be 5.62, 5.93 and 5.61units respectively. All the combinations showed increasing trend if frequency was increased except 70N grip force with stroke rotation 45°, grip force 90N with 45°and 60°.

Further three-way repeated measures ANOVA was performed on the standardised discomfort score (SDS), the results are presented in Table 3. Data was verified for order effect and found no significant different effect of order on discomfort score for all three groups of data of the discomfort score. The main effects frequency of exertion (p < 0.001) and Stroke rotation (p = 0.001) were significant on SDS (Table 3) but grip force was not significant on SDS. None of the interaction effects were significant on SDS.

The frequency and stroke rotation were found to be significant, therefore the post hoc Student Newman Keuls (SNK) tests were performed and the results were presented in Table 4 and Table 5. From Table 4 it was found that frequency 10 exertions/minute was significantly different from the frequency levels (15 and 20 exertions/minute) while frequencies 15 and 20 exertions/minute were found having same effect on SDS. As per Table 5 the stroke rotation 60° significantly different while 30° and 45° have the same effect on SDS.

Figure 3 showed the pattern of SDS for three levels of frequency of exertion at different stroke rotation angles. It was noticed that discomfort was higher for 20 exertions/minute than for remaining two frequencies at 30° stroke rotation. For both stroke rotations (45° and 60°), discomfort was found higher for 15 exertions/min than other two levels.

In Figure 4, the SDS for frequency 10 exertions/minute was found to be decreasing while increasing the level of grip force, maximum SDS was noted at 50N and minimum at 90N grip force.

Figure 5 showed that the SDS was increased with the increase in stroke rotation for frequencies 10 and 15 exertions/minute, while for 20 exertions/minute discomfort was found minimum at 45° than of the almost same values at 30° and 60°.

3.2 Electromyography (EMG) activities of forearm muscle

The RMS value for each of the condition was normalised for each muscle for each participant using the formula [41] as given below. $\begin{matrix} % Normalised EMG \\ = \frac{({rmsEMG}_{i} - {rmsEMG}_{min})}{({rmsEMG}_{\max} - {rmsEMG}_{min})} \times 100 % \end{matrix}$

The EMG raw data was transformed into useful and rmsEMG was normalised as in line with the studies [20 , 41]. The mean values of % normalised EMG and slope of median frequency of the signal are presented in Table 6. Using an attempt to identify which muscles were mostly affected in the respective combinations of the independent variables, using the mean values of % normalised EMG, was undertaken. Using the data, quartile plot was drawn and found that 25% of data shown of % Normalised EMG in Table 6 (value of % normalised EMG more than 12) was mostly affected as 22.15 was found from the data for muscle ECU with the particular combination (grip force 90N, 45° stroke rotation at 20 exertions/min); 20.45 was found for the same muscle at 60° stroke rotation with grip force 90N for 20 exertions/min; 20.05 for muscle FCU at 90N grip force with 45° stroke rotation for 20 exertions/min, for the torqueing with gripping task. Similarly, the Table 6 highlights data that was helpful for identifying the affectedmuscles for the particular combination of the experiment.

The values of the slope of median frequency were examined to assess whether muscles were relatively more fatigued for the respective combinations investigated in this experiment. A quartile plot was drawn using the data of Table 6 and found that 25% of the data (slope of MF more than –0.02 units) was highly fatigued for specific condition of the experiment. Muscles FCR, ECR and ECU were highly fatigued for different combinations as the value of slope of MF of muscle ECR at 50N grip force with 60° for 15 exertions/min was found to be (mean slope –0.06 units) which was the highest slope. Similarly, muscle ECU had the slope of Median Frequency (mean slope –0.06 units) for the combination; at grip force 70N with 45° stroke rotation for 10 exertions/min. The highlighted data of slope of Median Frequency displays which particular muscle was highly fatigued at specific combinations for the torqueing with the gripping task.

Table 7 summarizes the mean of % normalised EMG for individual levels of grip force, each levels of stroke rotation and each levels of frequency of exertions for individual forearm muscle. The maximum value of % normalised EMG was noticed for FCU (μ = 14.90), ECR and ECU at grip force 90N and the muscles FCR and FDS having the maximum at grip force 30N. The muscles were active during the repetitive forceful exertions in the supination rotation of forearm in the order, ECU, FCU, ECR, FCR and FDS for grip force 90N. Order of activeness of muscles was different for grip force 50N and 70N. All the muscles were affected during the task and increasing trend was found for the increases of frequency of exertions. All the muscles were most affected on 20 exertions/minute and least at 10 exertions/minute. The muscle ECU was highly affected due to frequency of exertions than other muscles. All the muscles were affected with the change in stroke rotation in some muscles the effect of stroke rotation was very high and in some low effect. Among all the muscles ECU affected most. The summary of % normalised EMG (Table 7) was found that the extensor muscles (μ = 11.82) were more affected than flexor muscles (μ = 9.63) in supination torqueing and gripping task.

Table 8 summarizes the summary data of slope and mean of median frequency during the task for individual level of grip force, each level of stroke rotation and for each level of frequency of exertion. For variable grip force, highest fatigue was found in the muscles FCR, FDS and ECU at 70N, the muscle FCU at 90N and ECR was found at 50N. At frequency 20 exertions/minute, muscle FCR and ECU were found to have high fatigue and at 10 exertions/minute muscles FCU, FDS and ECR have high fatigue in the said activity. All the muscles were fatigued due to an increase in frequency of exertions but extensor muscle (ECU) was highly fatigued in repetitive torqueing exertions. From the table of mean of median frequency, it was found that the highly fatigued muscle was ECU at all independent variables (grip force, frequency of exertions and stroke rotation).

The statistical analyses were carried out on the data of % normalised EMG. To begin the data of all variables related to EMG signals were verified and found no significant order effect on any parameter considered in MANOVA. Table 9 illustrated the results of MANOVA as there were more than one important parameters of EMG signals and found that the main effect of grip force was significant on % normalised EMG of muscles FCR (F_(2,216) = 6.941, p = 0.001), FCU (F_(2,216) = 8.589, p < 0.001) and ECR (F_(2,216) = 84.278, p = 0.008) and frequency of exertion was significant on muscles FCR (F_(2,216) = 3.655, p = 0.027), ECR (F_(2,216) = 13.709, p < 0.001) and ECU (F_(2,216) = 3.446, p = 0.034). Other main and interaction effects were not significant on % normalised EMG of any of the muscles. In general, flexor muscles were found significantly affected by grip force and extensor muscles were significantly affected by frequency of exertions in repetitive gripping and torqueing task.

Figure 6 illustrated that the % normalised EMG for muscle FCR was very low at grip force 70N as compared to 50N and slightly high at 90N. It was very surprising that the muscle FCU was highly activated at grip force 90N while there was no large difference in activation at grip force 50N as compared to 70N. The EMG activities of ECR muscle was found to be in between the muscle FCR and FCU at alllevels of grip force. A t-test was applied at grip force 90N and found that significant difference in activation of muscles FCR and FCU (t = –2.743, p = 0.008) and also significant difference in activation of muscles FCU compared to ECR (t = –2.552, p = 0.013). At 50N grip force t-test was applied and found that muscles FCR and FCU was significantly different (t = 5.943, p < 0.001) and muscle FCR and ECR was found to be significantly different (t = 3.226, p = 0.002). At grip force 70N One-Way Repeated measure ANOVA was applied and found that no significant difference of three muscles (FCR, FCU and ECR) on Normalised EMG (F_(2,160) = 1.975, p = 0.142).

Figure 7 illustrated that the percentage of normalised EMG of muscle ECU was seen to be high as compared to muscle FCR and ECR. Moreover percentage of normalised EMG for muscle FCR and muscle ECR was not very different for different levels of frequency of exertion. It was interestingly noticed that the lines of percentage normalised EMG for muscles FCR, ECR and ECU seems to be parallel, to confirm that parallelism test was applied on the data obtained for above said muscles at three levels of frequency.The differences of normalised EMG of each muscle at all level of frequencies were calculated and one-way repeated measure ANOVA was performed on the data and found that no significant difference between the difference of ECU and FCR muscles (p = 0.873) and muscles FCR and ECR (p = 0.129), showed that lines of three muscles were parallel.

Slope of median frequency (i.e. negative shift in power spectrum of the EMG signal) is an indicator of muscle fatigue. Therefore this parameter was also considered for statistical analyses. Histograms drawn for the values of the slope of median frequencies as well as for % normalised EMG showed closely normally distributed. The results of MANOVA performed (Table 9) on slope of median frequencies of different muscles as dependent variables showed that grip force was significant on the muscles FCR (p = 0.023) and ECU (p = 0.009); and stroke rotation was found to be significant on muscle FCU (0.015). No other main effect was found significant on slope of median frequency. The two way interaction of grip force and frequency was found significant on slope of median frequency of muscle ECU (0.028) and grip force with stroke rotation was significant on slope of median frequency of muscle ECU (0.012). A three way interaction of Grip Force and Frequency with Stroke Rotation) was significant on slope of median frequency of FCU (p = 0.019). It was observed from the findings of the Table 10 that in the change in the level of grip force made a significant effect on the EMG activities of the muscle FCR and ECU. However the levels of stroke rotation had significant effect on FCU muscle only. But the interaction effect of grip force with frequency and grip force with stroke rotation was on ECU muscle.

Figure 8 illustrated that the fatigue (mean slope –0.02 units) in muscle FCR was slightly increased at 70N grip force as compared to 50N. At a further increase in grip force (90N), the exertion in the muscle was noticed to be decreased. But it was noticed that for muscle ECU the fatigue (mean slope –0.03 units) at grip force 70N was much higher as compared to 50N and there was no further increase in fatigue at 90N grip force.

Figure 9 illustrated that the maximum fatigue (mean slope = –0.0102 units) in muscle FCU at 30° stroke rotation and found that fatigue was decreased as increased the level of stroke rotation. The minimum fatigue (mean slope –0.009 units) was found at 90° stroke rotation.

Figure 10 showed that the task performed at frequency of 10 exertions/minute, the muscle ECU was minimum fatigued at grip force 50N and continuously increased as grip force increases and having maximum fatigue (mean slope –0.031 units) at 90N grip force. At grip force 50N the muscle was found maximum fatigue for frequency of 15 exertions/minute while further increases in grip force the fatigue in muscle was found increased as compared to 50N for all levels of frequencies. Moreover the difference of fatigue in the muscle at 70N grip force was more as compared to the difference at 50N. Applied t-test on the data at grip force 70N for all levels of frequency and found that no significant difference between the data at 10 exertions/minute and 15 exertions/minute (p = 0.611) and 10 exertions/minute and 20 exertions/minute (p = 0.112). Applied t-test on the data of 90N grip force for all frequencies and found that no significant difference between frequency 10 exertions/minute & 15 exertions/minute (p = 0.051) and frequency 10 exertions/minute and 20 exertions/minute (p = 0.153).

Figure 11 illustrated that the minimum fatigue (mean slope –0.016 units) in muscle ECU was found at grip force 50N for all levels of frequency. It was noticed that the fatigue in a muscle was much more at 70N grip force and with further increases in grip force the fatigue in a muscle was approximately same as 70N for all levels of stroke rotation.

4 Discussion

4.1 Discomfort

4.1.1 Discomfort associated with grip force

Grip force was found not significant on discomfort in gripping repetitive task with torqueing in the present result. It was found that the decreases in discomfort as grip force increases. That might be due to different group of participants performed the task for a grip force so the perception of discomfort for participants was different. This was necessary to transform the raw data into standard data using the formula described above because some of previous studies [21, 37] have used standard discomfort score values. After transformation of data the result showed that the discomfort was highest for 70N grip force and the discomfort for 50N and 90N was slight same. That might be due to the reasons that the difference of level of grip force was very small (20N each), time of task was 5 min and physical characteristics (age, weight and height) as well as palm structure of all participants of each group were different, by which discomfort was felt differently. The present study differs from other studies in such a way that they applied % MVC grip force but this study performed the task at static grip force because in industries workers hold the tool with some constant force which depend on the type of task rather than subjective force of the participants. Previous studies [8 , 42] found that the grip force in % MVC was statistically significant on subjective discomfort rating during gripping exertions. O’Sullivan and Clancy [17] found that the given task’s force level (10N, 20N and 40N) was significant on discomfort in repetitive assembly task. Aldien et al. [43] concluded that grip forces (0, 15, 30, 50 and 75N) with push forces were strongly significant on the distribution of contact force in the hand-handle interface in gripping action. Marcotte et al. [44] reported that the grip force was strongly influenced on the biodynamic response of the hand–arm system in experiment based on grip force with push force. The findings of present study were not similar to other studies. That might be due to the use of the use of static grip force with only 20N difference from each level rather than % MVC.

4.1.2 Discomfort associated with stroke rotation

The stroke rotation was found significant on discomfort scores during the repetitive gripping exertions with torque in the present results. This was rotary action in supine direction for torqueing activity with grip, in which both flexor and extensor muscles were involved. It was difficult to directly relate the fatigue of the muscle with stroke rotation, however SNK test showed that 60° stroke rotation was at higher discomfort than other two stroke rotations. This might be due to more compression of the extensor muscles and extension of flexor muscles in supine rotation. EMG recording of the five muscles (FCR, FCU, FDP, ECR and ECU) showed that the effect of stroke rotation was not significant on % normalized EMG for any of the muscles but significant on slope of median frequency for FCU. Moment Arm (MA) explains biomechanical aspect of muscles movement, that possibly be reflected in terms of fatigue developed in muscles. Ettema et al. [45] and Gonzalez et al. [46] noticed that the Moment Arm (MA) was increased for the increased in forearm rotation angle from neutral to supine. They reported that the range of MA for FCR, FCU, ECR and ECU muscles was found 2.9 to 6.4, (–1.2) to 1.8, (–1.7) to (–0.7) and (–0.2) to 0.1 respectively for the 30°prone to supine rotation of forearm. These changes noticed in moment arms as per studies of [45, 46], might be the physiologicalreason of increased discomfort for the stroke rotation increase from 30° to 60°. Muscle architecture depends on moment arm, physiological cross sectional area, motor unit recruitment etc. Biomechanical potential for force is nothing but product of physiological cross sectional area and moment arm of muscle. Therefore it is understood that to get greater forcer with least effort more biomechanical potential is required. If the values obtained from the studies of [45, 46] were seen, it was found that biomechanical potential is generally increased with the increase in rotation angle of forearm in prone/supine direction for FCR, FCU, ECR, ECU muscles [47].

The present study differs from other studies because gripping exertions with dynamic posture of the forearm means continuous rotation of forearm while applying the force, but previous investigations performed the gripping exertions on the static position of the forearm. In some of the previous studies [8 , 25] reported that forearm angle significantly affected discomfort in repetitive task, although they have considered flexion force and torqueing tasks respectively. Mukhopadhyay et al. [8] considered isometric pronation torque with three forearm rotation angles and found that the discomfort level was significantly affected by forearm rotation angle also reported in Mukhopadhyay et al. [19, 24]. O’Sullivan and Gallwey [23] found that both forearm rotation angle and twisting directions significantly affected on discomfort for an intermittent pronation torque exertions.

4.1.3 Discomfort associated with frequency

Frequency of exertions found significant on discomfort in gripping with torqueing task. As SNK found that at 10 exertions/minute discomfort was lesser than other two frequencies and no significant difference between 15 and 20 exertions/minute. During EMG recording the extensor and FCR muscles were found more fatigue than other muscles due to elongation and compression in muscles during the supination task. In the literature, highly repetitive task may direct damage due to repeated stretching and elongation as well as fatigue increased and decrease chances of tissues to recover [48]. The objective measures of repetition include the use of kinematics. Kinematics metrics of the upper limb include calculating movement velocities (repetitiveness), joint deviations (posture) and forces exerted by muscles [17]. The movement velocities are the most significant parameters for identifying risks of injury [49]. The findings of the study were in-line of some studies [17, 18] performed repetitive task with repetitions 10, 15 and 20 exertions/minute and found that the repetitions were significant on discomfort. Khan et al. [27] concluded the frequency (10 and 20 exertions/minute) was highly significant on discomfort in repetitive wrist flexion task. Ciriello et al. [31] conducted the screw driving task using repetition rates 15, 20 and 25 motions/min and reported that the maximum isometric torque was highly depend on frequency.

4.2 Electromyography

EMG is widely used as a dependent variable to correlate feeling of the discomfort with the WMSDs, may be developed due to postural, force and frequency related factors [8 , 51]. In general the analysis showed that the extension muscles were more fatigued compared to flexion muscles for the forearm. Although in this study only five superficial muscles of the forearm were recruited for the EMG recording. These muscles by and large give the general idea about the fatigue of the forearm as many researchers have used them for the evaluation of repetitive tasks in simulation [8 , 53] or occupational environment [54 –56]. In line with the results of the present study O’Sullivan and Gallwey [21] has reported highly significant effect of pronation and supination torque on ECRB muscle. Hoozemans and Dieën [35] also reported that extensor muscle activity is highly associated with the power grip activity. The task of the present study was a combination of power grip and supination torque and has found support that extensor muscles are more affected for gripping combined with torque for repetitive exertions.

Slope of the median frequency indicative of the muscle showed that grip force levels 70 and 90N had significantly faster rate of fatigue development compared to the force of 50N. This showed that the task below the level of 70N grip could be prolonged for a longer duration at the frequency range 10-20 exertions/minute. Percentage Normalised EMG was around 12% in the extensor muscles compared to approx. 8% for flexion muscles (Table 6). It also found that for gripping tasks, the fatigue in extensors was found more quickly than flexor muscles [57, 58]. These outcomes are in line with the previous studies [34 , 59] those reported higher muscle activity for tasks involving grip. It is known that muscle force and muscle length play an important role in defining the relationship between EMG amplitude and grip force. Muscle length changes with the rotation of forearm [60], changes in muscle arm occur due to change in muscle length [61] and change in moment arm with the change in forearm rotation [45, 46]. These valuesof the MA indicating changes relative to forearmrotation were presented above.

5 Limitations

There were following possible limitations of the findings if applied in industrial task or tool design:

The duration of the experiment was very short as 5 minutes, hence findings may be helpful to tasks having shorter cycles times.

The range of grip force levels investigated was only at a part 20N.

Due to some cultural bindings here it was difficult to have female participants.

In this study task was given of one type and the other important limitation was that wrist, elbow and shoulder angles were considered. This may be taken for further investigations in continuation of these findings.

6 Conclusions

Frequency of exertions was highly significant on (SDS) and found that torqueing combined with grip was more discomforting at levels of 15 and 20 exertions/minute. The 30° and 45° stroke rotation had no significant difference on SDS however 60° stroke rotation was significantly difference from both and higher then both. It was found that stroke rotation for the torqueing tasks must be kept below 45°. Extensor muscles were found to be highly active and more fatigued compared to flexor muscles for the experimental task especially ECU muscle. Also it was found that more % normalised EMG were noticed for 15 and 20 exertions/minute compared to 10 exertions per minute, indicating the similar trend of the SDS. The findings of the present study will help task designers to consider requirement of forearm rotation in different types of tasks requiring grip and toque all together. This will also be helpful in tools design such as plier, screw driver etc., wherever stroke rotation and grip force have important significance.

Footnotes

Acknowledgments

Authors are thankful to the funding agencies, in result of this a good research facility has been developed. That was very helpful to perform these experiments.

All India Council of Technical Education (AICTE, New Delhi) Reference No. F/No 8023/BOR/RID/RPS- 23/2008-09.

DST, MHRD, Fast Track – SERC Scheme No.SR/FTP/ETA-64/2009/dated Dec 2, 2009.

References

Fan

, Silverstein

, Bao

, Bonauto

, Howard

, Spielholz

, Viikari-Juntura

. Quantitative exposure-response relations between physical workload and prevalence of lateral epicondylitis in a working population. Am J Ind Med 2009;52(6):479–90.

Harris

, Eisen

, Goldberg

, Krause

, Rempel

. 1st place, PREMUS best paper competition: Workplace and individual factors in wrist tendinosis among blue-collar workers. The San Francisco study. Scand J Work Environ Health 2011;37(2):85–98.

Silverstein

, Fan

, Bonauto

, Bao

, Smith

, Howard

, et al. The natural course of carpal tunnel syndrome in a working population. Scand J Work Environ Health 2010;36(5):384–93.

Bernard

. Musculoskeletal disorders and workplace factors: Cincinnati: National Institute for Occupational Safety and Health; 1997.

Kumar

. Theories of musculoskeletal injury causation. Ergonomics 2001;44(1):17–47.

Wiker

, Chaffin

, Langolf

. Shoulder posture and localized muscle fatigue and discomfort. Ergonomics 1989;32:211–37.

Bureau of Labor Statistics (USA), 2004. Lost-work time injuries and illnesses. Characteristics and resulting days away from work, 2002. U.S. Department of Labor.

Mukhopadhyay

, O’Sullivan

, Gallwey

. Estimating upper limb discomfort level due to intermittent isometric pronation torque with various combinations of elbow angles, forearm rotation angles, force and frequency with upper arm at 90° abduction. Int J Ind Ergon 2007a;37(4):313–25.

Hughes

, Silverstein

, Evanoff

. Risk factors for work-related musculoskeletal disorders in an aluminium smelter. Am J Ind Med 1997;32:66–75.

10.

Herbert

, Dropkin

, Warren

, et al. Impact of joint labor-management ergonomics program on upper extremity musculoskeletal symptoms among garment workers. Appl Ergon 2001;32:453–60.

11.

Hagberg

, Wegman

. Prevalence rates and odds ratios of shoulder-neck diseases in different occupational groups. Occup Environ Med 1987;44(9):602–10.

12.

Centers for Disease Control and Prevention. (2011). Workplace Health Promotion. Retrieved August 13, 2012, from http://www.cdc.gov/workplacehealthpromotion/evaluation/topics/disorders.html

13.

Armstrong

, Radwin

, Hansen

, Kennedy

. Repetitive trauma disorders: Job evaluation and design. Human Factors 1986;28(3):325–36.

14.

Moore

, Wells

, Ranney

. Quantifying exposure in occup[ational manual tasks with cumulative trauma disorder potential. Ergonomics 1991;34:1433–53.

15.

, Yu

. Assessment of grip force and subjective hand force exertion under handedness and postural conditions. Appl Ergon 2011;42(6):929–33.

16.

Lin

, McGorry

. Predicting subjective perceptions of powered tool torque reactions. Appl Ergon 2009;40(1):47–55.

17.

O’Sullivan

, Clancy

. Guideline threshold limit values (TLVs) for discomfort in repetitive assembly work. HumFactors Ergon Manuf 2007;17(5):423–34.

18.

Finneran

, O’Sullivan

. Force, posture and repetition induced discomfort as a mediator in self-paced cycle time. Int J Ind Ergon 2010;40(3):257–66.

19.

Mukhopadhyay

, O’Sullivan

, Gallwey

. Upper limb discomfort profile due to intermittent isometric pronation torque at different postural combinations of the shoulder-arm system. Ergonomics 2009;52(5):584–600.

20.

Mehta

, Agnew

. Analysis of individual and occupational risk factors on task performance and biomechanical demands for a simulated drilling task. Int J Ind Ergon 2010;40(5):584–91.

21.

O’Sullivan

, Gallwey

. Upper-limb surface electro-myography at maximum supination and pronation torques: The effect of elbow and forearm angle. J Electromyogr Kinesiol 2002;12(4):275–85.

22.

Dimberg

. The prevalence and causation of tennis elbow (lateral epicondylitis) in a population of workers in an engineering industry. Ergonomics 1987;30:573–80.

23.

O’Sullivan

, Gallwey

. Forearm torque strengths and discomfort profiles in pronation and supination. Ergonomics 2005;48:703–21.

24.

Mukhopadhyay

, O’Sullivan

, Gallwey

. Effects of upper arm articulations on shoulder- arm discomfort profile in a pronation task. Occupational Ergonomics 2007b;7:1–13.

25.

Khan

, O’Sullivan

, Gallwey

. Effects of combined wrist flexion/extension and forearm rotation and two levels of relative force on discomfort. Ergonomics 2009a;52(10):1265–75.

26.

Khan

, O’Sullivan

, Gallwey

. Effects of combined wrist deviation and forearm rotation on discomfort score. Ergonomics 2009b;52(3):345–61.

27.

Khan

, O’Sullivan

, Gallwey

. Effect on discomfort of frequency of wrist exertions combined with wrist articulations and forearm rotation. Int J Ind Ergon 2010;40(5):492–503.

28.

Farooq

, Khan

. Effect of shoulder rotation, upper arm rotation and elbow flexion in a repetitive gripping task. Work 2012;43(3):263–78.

29.

Chang

, Wang

MJJ

. Evaluating factors that influence hand- arm stress while operating an electric screwdriver. Appl Ergon 2000;31:283–9.

30.

Chang

, Wang

MJJ

. Evaluating the effects of activation mode, torque and horizontal operating distance on hand-arm response while operating pneumatic screwdrivers. Int J Ind Ergon 2001;28:171–9.

31.

Ciriello

, Webster

, Dempsey

. Maximal Acceptable Torques of Highly Repetitive Screw Driving, Ulnar Deviation, and Handgrip Tasks for 7-Hour Workdays. AIHA J 2002;63(5):594–604.

32.

Mukhopadhyay

, Gallwey

, O’Sullivan

. Effect of grip force and shoulder abduction angles in repetitive clockwise forearm pronation/supination. In: Proceedings of the International Annual Occupational Ergonomics and Safety Conference. 2003.

33.

Bano

, Mallick

, Khan

. Effect of grip, stroke rotation and handle size on discomfort for screwing task. Int J Hum Factors Ergon 2012;1(4):390.

34.

Eksioglu

. Endurance time of grip-force as a function of grip-span, posture and anthropometric variables. Int J Ind Ergon 2011;41(5):401–9.

35.

Hoozemans

MJM

, van Dieën

. Prediction of handgrip forces using surface EMG of forearm muscles. J Electromyogr Kinesiol 2005;15(4):358–66.

36.

Mogk

JPM

, Keir

. The effects of posture on forearm muscle loading during gripping. Ergonomics 2003;46(9):956–75.

37.

Carey

, Gallwey

. Wrist discomfort levels for combined movements at constant force and repetition rate. Ergonomics 2005;48(2):171–86.

38.

Lin

, Radwin

, Snook

. A single metric for quantifying biomechanical stress in repetitive motions and exertions. Ergonomics 1997;40(5):543–58.

39.

Carey

, Gallwey

. Effects of wrist posture, pace and exertion on discomfort. Int J Ind Ergon 2002;29:85–94.

40.

Gescheider

, editor. Psychophysics – Method, theory, and application. 2nd ed. Hillsdale (NJ): Lawrence Earlbaum; 1985.

41.

Strasser

. Electromyography: Methods and techniques. In: Karwowski

, editors. International Encyclopaedia of Ergonomics and Human factors. London: Taylor and Francis; 2001;3:1801–4.

42.

Kong

Y-K

, Kim

D-M

, Lee

K-S

, Jung

M-C

. Comparison of comfort, discomfort, and continuum ratings of force levels and hand regions during gripping exertions. Appl Ergon 2012;43(2):283–9.

43.

Aldien

, Welcome

, Rakheja

, Dong

, Boileau

P-E

. Contact pressure distribution at hand– handle interface: Role of hand forces and handle size. Int J Ind Ergon 2005;35(3):267–86.

44.

Marcotte

, Aldien

, Boileau

, Rakheja

, Boutin

. Effect of handle size and hand– handle contact force on the biodynamic response of the hand–arm system under zh-axis vibration. J Sound Vib 2005;283:1071–91.

45.

Ettema

, Styles

, Kippers

. The moment arms of 23 muscle segments of the upper limb with varying elbow and forearm positions: Implications for motor control. Hum Mov Sci 1998;17(2):201–20.

46.

Gonzalez

, Buchanan

, Delp

. How muscle architecture and moment arms affect wrist flexion-extension moments. J Biomech 1997;30(7):705–12.

47.

Khan

. Effect of wrist articulation combined with forearm rotation on discomfort, Ph.D. Thesis, University of Limerick, Ireland. 2004.

48.

Armstrong

. Ergonomics and cumulative trauma disorders of the hand and wrist. In: Hunter

, Schneider

, Mackin

, Callahan

editors. Rehabilitation of the hand surgery and therapy. Philadelphia: Mosby; 1990.

49.

Marras

, Schoenmarklin

. Quantification of wrist motion in highly repetitive, hand intensive industrial jobs. Final Report. Grant nos. 1 R01 OH02621-01 and 02 funded by NIOSH. 1991.

50.

Attebrant

, Winkel

, Mathiassen

, Kjellberg

. Shoulder-arm muscle load and performance during control operation in forestry machines. Appl Ergon 1997;28(2):85–97.

51.

Buchholz

, Park

J-S

, Gold

, Punnett

. Subjective ratings of upper extremity exposures: Inter-method agreement with direct measurement of exposures. Ergonomics 2008;51(7):1064–77.

52.

Tomatis

, Nakaseko

, Läubli

. Co-activation and maxima EMG activity of forearm muscles during key tapping. Int J Ind Ergon 2009;39:749–55.

53.

Johnson

, Ciriello

, Kerin

, Dennerlein

. Using electrical stimulation to measure physiological changes in the human extensor carpi ulnaris muscle after prolonged low-level repetitive ulnar deviation. Appl Ergon 2013;44(1):35–41.

54.

Sjøgaard

, Chriestensen

, Pedersen

. Neck and shoulder disorders due to prolonged uncomfortable working postures. In Proc. of Eleventh Congress of the International Ergonomics 1991;1: 123–25. Paris.

55.

. Ergonomic design and evaluation of wire-tying hand tools. Int J Int Ergon 2002;30:149–61.

56.

. Ergonomic evaluation of a fixture used for power driven wire-tying hand tools. Int J Int Ergon 2003;32:71–79.

57.

Byström

SEG

, Mathiassen

, Fransson-Hall

. Physiological effects of micropauses in isometric handgrip exercise. Eur J Appl Physiol Occup Physiol 1991;63(6):405–11.

58.

Hägg

, Milerad

. Forearm extensor and flexor muscle exertion during simulated gripping work — an electromyographic study. Clin Biomech 1997;12(1):39–43.

59.

Roman-Liu

, Tokarski

. EMG of arm and forearm muscle activities with regards to handgrip force in relation to upper limb location. Acta Bioeng Biomech 2002;4(2):33–44.

60.

Ljung

B-O

, Fridén

, Lieber

. Sarcomere length varies with wrist ulnar deviation but not forearm pronation in the extensor carpi radialis brevis muscle. J Biomech 1999;32(2):199–202.

61.

Loren

, Shoemaker

, Burkholder

, Jacobson

, Fridán

, Lieber

. Human wrist motors: Biomechanical design and application to tendon transfers. J Biomech 1996;29(3):331–42.