Correlation between muscle strength, fatigue resistance, work ability and upper limb dysfunction in a sample of workers at a university hospital


 IntroductionThe upper extremity is related to the functional ability to perform activities of daily living, self-care and work. Complaints in the upper limbs negatively affect these skills, therefore, it is necessary to obtain information about work demands.ObjectiveTo correlate upper limb muscle strength with fatigue resistance, work ability and dysfunction in a sample of workers from a university hospital.MethodsShoulder and elbow strength were assessed by Biodex System 4™, hand grip by JAMAR™, fatigue resistance by Functional Impairment test Hand/Neck/Shoulder/Arm (FIT-HaNSA), the ability to work by the Work Ability Index and upper limb dysfunction by QUICK-DASH-Br. The Brazilian version of the Pain Catastrophising Scale (B-PCS) was applied in order to analyse the catastrophic profile of the sample. The correlations were analysed by Spearman's Correlation Coefficient (rho)(p < 0.05).Results27 workers presented with chronic pain of low intensity, were classified as active and underwent predominantly dynamic work. Muscle strength of abduction (0.49), adduction (0.40), internal rotation (0.44), elbow flexion (0.38) and hand grip (0.68) had a direct correlation with FIT-HaNSA. Only hand grip (-0.52) showed an inverse correlation with upper limb dysfunction. Muscle strength was not correlated with work ability or pain catastrophising.ConclusionThe results suggested that increasing muscle strength may improve fatigue resistance. Also, improving hand grip strength could decrease upper limb dysfunction. In contrast to previous findings, muscle strength was not related to the ability to work or to pain catastrophising, because these factors are not only influenced by physical function, but also by psychosocial functioning.


Introduction
The upper extremity is related to the functional capacity to perform activities of daily living (ADLs), selfcare tasks and sports (Soyer et al., 2016). When a person experiences pain and reduced mobility in this segment, functional skills, work performance and quality of life are negatively affected (Schunid et al., 2003).

Page 3/18
Physical demands involving repetitive work, lack of muscle recovery, precision of movements and static postures are risk factors for the development of upper limb disorders (Nordander et al., 2009;Sundstrup et al., 2014;Ricco and Signorelli, 2017;Subramaniam et al., 2018). In particular, when associated with psychosocial factors, musculoskeletal disorders in the upper limbs generate high costs, long-term sickness absence and decreased productivity in many economic sectors (Lomond and Coté, 2009;Nordander et al., 2009;Eerd et al., 2016).
Since there is a substantial prevalence (36.8%) for complaints in the upper limbs, it is useful to obtain accurate information about work demands related to this segment (Jacobs et al., 2017). Thus, rehabilitation or prevention programs, ergonomic assessment and education strategies must be considered (Williams and Westmorland, 1994).
A careful musculoskeletal assessment should involve the analysis of all variables associated with the individual's health, such as the level of activity and function, characteristics of symptoms and level of strength to perform daily activities and work (Skirven, et al., 2011).
Muscle strength is the greatest predictor of function, mobility and independence (Roshanravan et al., 2016). An instrument used for this measurement is the dynamometer, which can be either isometric or isokinetic. Isokinetic evaluation analyses strength, power, work, fatigue percentage and muscle performance, and classi es these data as normal or altered. Thus, it allows the evaluation of the effectiveness of a treatment, dictates the rehabilitation objectives and establishes normative values of In addition to strength, fatigue resistance is another indicator of functionality and physical capacity. Work demand is an important ability to be analysed, considering that fatigue is a complaint frequently reported among workers. In this context, it can be de ned as an imbalance between work demands and the worker's ability to perform a task. It is associated with a decrease in motor performance, speed and range of motion, and may be caused by repetition of a single activity for prolonged periods. Long-term fatigue increases the risk of incapacity for work and impairs the ability to maintain a satisfactory level of To gain a better understanding of the variables involved in workers' musculoskeletal complaints, the aim of this study was to analyse the correlations between the muscle strength of the shoulder, elbow and hand with fatigue resistance, work ability and upper limb dysfunction in a sample of workers at a university hospital.

Method
This was a cross-sectional observational study which was approved by the local Research Ethics Committee (No. 2,724,782, CAAE 89138818.1.0000.5440) in accordance with the 1964 Helsinki Declaration and its subsequent amendments. All participants signed an informed consent form before participating in the study.
Sample The sample size calculation was performed with alpha 0.05, power 0.8 and an effect size of 10%, and identi ed 25 workers that were selected from different sectors at a tertiary-level university hospital. The included participants were individuals of both genders, aged between 25 and 60 years, who had pain or discomfort in the upper limb over the previous 12 months, which may have been unilateral or bilateral, and who predominantly used the upper limbs for work. We excluded people who had recent surgery or trauma and were unable to perform the proposed tests for physical or cognitive reasons. OutcomesNumerical Pain Rating Scale (NPRS) Using an 11-point scale ranging from zero (no pain) to ten (worst pain ever), participants were instructed to choose the numerical value that best represented their pain intensity at that moment, and a rating was made as follows: 0 (zero) -no pain, from 01 to 03 -mild pain, from 04 to 06 -moderate pain and from 07 to 10 -severe pain (Hartrick et al., 2003).
Nordic Musculoskeletal Questionnaire This was used to de ne the location and chronicity of symptoms. It consisted of a body map divided into 09 segments: neck, shoulders, upper back, elbows, wrists/hands, lower back, hips/thighs, knees and ankles/feet, and presented four questions with binary responses ("Yes/no") for each segment (Pinheiro, 2002). International Physical Activity Questionnaire (IPAQ) -Short Version This instrument was applied in order to estimate the weekly time spent in physical activities carried out in a normal week, taking into account walking activities, activities with moderate and vigorous intensity that had a minimum duration of 10 continuous minutes, in addition to the time spent in the sitting position.
The participants were characterised as very active, active or irregularly active and sedentary (Matsudo et al., 2001).

Quick-DASH-Br
This questionnaire was used to de ne the upper limb dysfunction level. It was composed of 11 items, taken from the DASH questionnaire of 30 items, in order to assess the symptoms, and physical and social function related to complaints in the upper limb. Each item was scored on a scale from 01 to 05 points, where 01 indicated "no di culty" and 05 "extreme di culty", with a nal score ranging from 0 to

Work Ability Index
This instrument was used to assess, through the worker's self-report, the participants' perception of working conditions and physical, mental and social capacity that may be related to their complaints. The index contained 10 questions divided into seven domains and the results provided a measure of work ability, which ranged from 07 to 49 points, and classi ed the ability as low (07 to 27), moderate (28 to 36), good (37 to 43) or excellent (44 to 49) (Martinez et al., 2009).
Pain Catastrophising Scale (B-PCS) This scale was used to assess the degree of pain catastrophising. The scale contained 13 items, in which the participant had to indicate the degree to which he/she presented any thought or feeling described, in a 05-point graduation: Minimum (0), Light (1), Moderate (2), Intense (3) or Very Intense (4). The total score was calculated by the sum of all items, ranging from 0 to 52, indicating the catastrophising level as high (> 30) or low (≤ 30) (Sullivan, 2009;Mosegaard et al., 2020). Isokinetic and Isometric Dynamometer Muscle strength was assessed using the Biodex System 4 Pro™ isokinetic dynamometer following all the calibration and use recommendations proposed by the manufacturer in the manual. The force variable used was the mean torque peak, at a speed of 60º/sec, as recommended for the evaluation of orthopaedic complaints (Prentice and Voight, 2003). The participant's position was based on the guidance material provided by the BiodexÔ system. The movements evaluated were abduction/adduction of the shoulder in the scapular plane, based on the angle proposed by Kapandji (2000), positioning the shoulder at 30º anterior to the frontal plane (Fig 01.), internal/external rotation of the shoulder (Fig 02.) and elbow exion and extension (Fig 03). Fatigue resistance A JobSim™ System prototype was developed and used to perform the fatigue resistance test "Functional Impairment Test-Hand, and Neck/Shoulder/Arm" (FIT-HaNSA). It is a three-task test that represents upper limb gross motor functions, such as reaching and holding objects at different heights and work sustained above the head (MACDERMID et al., 2007) (Fig 05.). The test consisted of performing two unimanual and one bimanual task, developed in the following steps: Task 1) Waist-up: One shelf was placed at waist level and the other 25 cm above it. Three 01kg containers were placed on the lowest shelf. The participant was instructed to move the three containers from one shelf to another, at a speed of 60 beats/minute, controlled by a metronome.
Task 2) Eye-Down: One shelf was placed at eye level and the other 25 cm below it. The participant was again instructed to move the three containers from one shelf to another at the same speed.
Task 3) Overhead-Work: A board containing screws was attached perpendicularly to the shelf at eye level. The participants were instructed to keep both arms raised and use them to screw and unscrew the screws in a predetermined sequence: the screw that was on level 01 (top) must be moved to level 02 (middle); the screw from level 03 (lower) to level 01 and the screw from level 02 to level 03.
Each task was performed only once for a maximum of 300 seconds or when the volunteer interrupted the test due to pain, fatigue or disability. The rest between one task and another was established by the time taken to adjust the shelf for the subsequent task (MacDermid et al., 2007). To analyse the correlations, the average time of the three tasks was calculated.

Procedure
The evaluation was carried out in a single day and applied by the same physiotherapist. If the participant had bilateral complaints or pain in more than one joint of the upper limb, the site and member of highest pain intensity was de ned as the symptomatic member/site measured by NPRS. Before starting the tests, instrument familiarisation and randomisation of muscle strength and fatigue resistance tests was performed. The tests were preceded by a 05-minute warm up on the cycle ergometer and as a rest criterion, a socio-demographic form and the questionnaires were applied.

Statistical Analysis
The correlations were analysed by Spearman's Correlation Coe cient (rho) which measures the degree of linear association between two variables. This coe cient is a measure used in non-parametric samples, for two ordinal variables or when one variable is continuous and the other is categorical or non-normal (Di

Results
Twenty-six out of twenty-seven participants were right-handed and only one was left-handed, with handedness being assessed by The Lateral Preference Inventory (Marim et al., 2011). 59.25% reported pain in the dominant limb and 40.74% in the non-dominant limb. All of the participants predominantly used the upper limbs in their functions but with different levels of effort. 62.96% performed dynamic work, with a high level of physical effort, including lifting/shifting loads; 11.11% performed dynamic work but without lifting/shifting loads; and 25.92% had static work characteristics with a low level of physical effort, and maintaining the same position for prolonged periods. The selection of participants and work sectors are described in the owchart in Fig. 06.
There was a similar proportion of men and women aged between 26 and 59 years. By IPAQ questionnaire, 18.51% were classi ed as very active, 37.03% as active, 37.03% as irregularly active and 7.40% as sedentary. Data that characterise the sample are summarised in table 01. Pain intensity was classi ed as mild and the Quick-DASH-Br showed that workers had a low level of upper limb dysfunction. A de cit of 50% in fatigue resistance was found in the symptomatic limb when comparing the expected total time with the total time spent to perform the three tasks. These values are presented in table 02. Muscle strength showed a signi cant correlation with FIT-HaNSA, except in regard to external shoulder rotation and elbow extension, with direct and weak correlations for abduction, adduction, internal rotation and elbow exion movements and a moderate correlation for hand grip. Hand grip strength also showed a moderate inverse correlation with upper limb dysfunction. The correlations between the other proposed constructs were not signi cant. Table 03 shows the numbers of correlations obtained and the respective statistical signi cance level, with a 95% con dence interval (p < 0.05). Graphic 01 shows the average strength values obtained for women and men.  These studies agree with our data sample, since more than half reported long-term sickness absence due to musculoskeletal complaints and complained of pain in other segments besides the upper limbs.
Fatigue resistance was impaired in this sample. Despite the weak magnitude of correlation, our data showed a signi cant and direct interaction between fatigue resistance and muscle strength for all moviments, except for the external shoulder rotation and elbow extension, suggesting that muscle strength and fatigue resistance change in the same direction. Sundstrup and collaborators (2014) outlined that musculoskeletal pain aggravates the development of muscle fatigue. They observed that resistance training signi cantly improved muscle strength and reduced pain and disability, suggesting an interaction between fatigue resistance and strength in a sample of workers complaining of chronic pain in the upper limbs.
It is known that hand grip strength is correlated with different functional disorders of the upper limbs, such as rheumatoid arthritis, carpal tunnel syndrome, lateral epicondylalgia, stroke, traumatic injuries and neuromuscular diseases, and it is also strongly related to general health status (Novaes et al., 2009;HORSLEY, et al., 2016). Our study agrees with these ndings, as we observed a signi cant and moderate inverse correlation between hand grip strength and upper limb dysfunction, showing that the dysfunction decreased when the hand grip strength increased.
Even with a low level of upper limb dysfunction and a satisfactory level of physical activity, these characteristics were not enough for most participants to perceive that they had an excellent work ability, since the ability to work was considered moderate. There was no correlation between muscle strength and ability to work. This can be explained by the fact that the ability to work is not only in uenced by physical function, but also by psychological, cognitive and social functioning. These factors are, in turn, affected by pain, and the association of these variables negatively interferes with the ability to work (Hengstebeck et al., 2017).
Regarding psychosocial issues, only 35.29% recorded a high catastrophising score, and this score did not

Study Limitations
The methodology used does not allow us to establish a causal factor or risk factors, so the ndings are only suggestive of interactions between the constructs. In addition, the sample size did not allow the analysis of subgroups that considered the individuality of each worker pro le in terms of magnitude and characteristics of effort.
There were also no detailed analyses on the ergonomic conditions of the work environment nor on the social and psychological relationships of the workers, which would be relevant considering that the ability to work could be in uenced by these aspects.

Conclusion
The correlation between muscle strength and fatigue resistance suggests that increasing muscle strength may improve fatigue resistance in workers with upper limb complaints.
Improving hand grip strength can be important to decrease upper limb dysfunction in workers since these variables showed an inverse correlation.
In contrast to previous ndings, muscle strength was not related to the ability to work or to pain catastrophising, and it is likely that this is because these factors are not only in uenced by physical functioning, but also by psychological, cognitive and social functioning.
These ndings may be useful for future analyses of similar worker pro les, and may have particular relevance for the development of interventions aimed at preventing the development of musculoskeletal complaints in the upper limbs related to work.

Declarations
• Ethics approval and consent to participate     Values of strength for women and men