Association of maximal voluntary isometric handgrip strength with age , gender and handedness in older people

Background: Maximal voluntary isometric handgrip strength (MVIHS) is influenced by age, sex, and handedness. Aim: To assess the association of MVIHS with age, sex, and handedness in older adults. Material and Methods: MVIHS was measured using a digital dynamometer in 60 men and 60 women aged 73 ± 6 years. Weight, height and handedness were also recorded. For analysis purposes, participants were divided into two age groups (65 to 70.9 years of age and ≥ 71 years). Results: A negative correlation was observed between age and MVIHS in the non-dominant (r = -0.65 and -0.59 in men and women, respectively) and dominant hands (r = -0.71 and -0.64 in men and women, respectively). When age and MVIHS were correlated in the group aged 65-70 years, a significant correlation was observed in the non-dominant (r = -045 and -0.61 in men and women, respectively) and dominant hands (r = -0.47 and -0.64 in men and women, respectively). In the group aged ≥ 71 years, a stronger correlation with age was also observed in the non-dominant (r = -0.92 and -0.90 in men and women, respectively) and dominant hands (r = -0.95 and -0.90 in men and women, respectively). MVIHS was 2.8 to 8.9% lower in the non-dominant than in the dominant hand in all age groups. MVIHS was lower in women than in men in both age groups. Conclusions: MVIHS declines with age (especially after 71 years of age), is higher in men than women, and higher in the dominant than the non-dominant hand. (Rev Med Chile 2018; 146: 1429-1437)


Material y Métodos: La FAIVM, la masa corporal, la talla, y la dominancia fueron medidas mediante protocolos estandarizados en 60 hombres e igual número de mujeres que fueron divididos en dos grupos acorde a su edad (65 a
Handgrip strength in older adults -J.Hernández-Martínez et al Rev Med Chile 2018; 146: 1429-1437 C hile shows an aging pattern and for the year 2025 it is expected that 20.1% of its adult population will be constituted by older adults 1 .Aging is a complex natural process, implicating psychological and social changes, alongside with physical changes, with a potential negative impact on quality of life 2 .A key issue is the progressive reduction of physical fitness along with a reduction in muscle strength (dynapenia) 3,4 .Dynapenia may occur with independence from neurologic or muscular diseases 4 and is associated with the progressive dysfunction of the endocrine, neural, cardiovascular, and musculoskeletal systems, where the reduction of muscle mass, negative changes in mental health, nutritional status, physical inactivity and sedentary behavior contributes to increase the rate of dynapenia [4][5][6] .This loss of muscle strength negatively affects physical performance and mobility, leading to increased risk of functional dependency, disability, and morbimortality among older adults [6][7][8] .Therefore, the opportune muscle-strength assessment status of older adults is a key issue, ideally with simple, inexpensive and reliable tests 8 .
Maximal isometric dynamometric handgrip strength is a simple and inexpensive test to assess muscle strength, with high reliability and validity [9][10][11][12] to estimate the functionality in older adults due to their elevated association with the strength and mass of several muscle groups 8 .Also, handgrip strength in older adults is associated with independent-living, quality of life 13 , health 14 , may help to identify population at risk of functional limitations 15 , frailty 16 , falling, morbidity after sur-gery, among others relevant aspects 17 .In addition, a recent study showed a significant association between handgrip strength and several cardiovascular risk factors in Chilean older population 18 .
Internationally populations-based handgrip strength values usually are expressed according to age and sex, while the relationship with handedness has demonstrated contrasting results 19,20 .Usually men exhibit greater handgrip strength compared to women 12,14,20 and maximal values are achieved during the fourth decade of life 21 , and from there a strong inverse relationship exists between handgrip strength and age 22,23 .Regarding handedness, right-handed subjects are stronger with their right hand, while left-handed subjects show unclear results 24 .
In Chile, reports regarding handgrip strength in older populations are scarce [25][26][27] , usually employing different measurement equipment and protocols.Considering the important differences between studied populations and methodologies employed for determination of the handgrip strength in older adults, both at internationally [28][29][30] and national level [25][26][27] , critical limitations have emerged regarding its clinical applicability in community-health programs for older adults in Chile, being necessary to count with standardized reference values in view of its potential applicability in public health policies.

Subjects and procedures
A descriptive, exploratory, cross-sectional study was conducted with community-dwelling older participants of Hispanic descent from an urban area in a southern city in Chile, in the commune of Osorno.The study was conducted in the Laboratory of Human Performance, at the University of Los Lagos.Inclusion criteria required i) being free of dependency risk according to the Functional Diagnosis for Older Adults from the Preventive Medical Exam for Older Adults (EM-PAM) by the Chilean Ministry of Health 35 , ii) having non-altered punctuation in the Mini Mental Status Examination test (short form) according to Chilean standards from the EMPAM 35 , iii) being ≥65 years of age, iv) being free of risk according to the Physical Activity Readiness Questionary (PAR-Q) for older adults 36 .Excluded subjects were those with i) blindness, severe hearing loss, upper limb amputation, stroke-related complications or any physical or mental condition that could affect the results of dependent variables.According to these requirements, from 184 participants initially screened only 120 (female, n = 60; male, n = 60) where included in the study.All subjects were carefully informed about the experimental procedures and about the possible risks and benefits associated with participation in the study and signed an informed consent document before any of the tests were performed.The study was conducted in accordance with the Declaration of Helsinki and was approved by the ethics committee of the Department of Physical Activity Sciences, from the University of Los Lagos.
Standing height (m) and body mass (kg) were assessed according to international standards for anthropometric assessment 37 .Anthropometry comprised stature on a stadiometer (Bodymeter 206, SECA, Germany) and body mass on an electrical scale (BF 100_Body Complete, Beurer, Germany), with precisions of 0.1 cm and 0.1 kg, respectively.These parameters were assessed prior to the MVIHS test.Subjects were tested while wearing light clothing (shoes were removed).The body mass index (BMI) was calculated (kg/m -2 ).

Measurement of handgrip strength
Participants were required to avoid unusual physical activities at least 24 hours before MVIHS measurements, with especial emphasis in activities that may have implicated an unusual high-load for the upper extremities.This requirement was verified with a personal interview at the beginning of the measurement day.To assess MVIHS an adjustable digital dynamometer was used (Jamar®, PLUS+, Sammons Preston, Patterson Medical, Illinois, United States).After randomly assigning the order of dominant and non-dominant hand assessment (i.e., handedness was assesses based on previous recommendations) 38 , subjects completed a standardized general and specific familiarization warm-up protocol, consisting of ten submaximal contractions of the forearm muscles (using a squeeze-ball) and three submaximal handgrip attempts (using the digital dynamometer).Then, subjects had three valid attempts to achieve MVI-HS for both dominant and non-dominant hands, with two minutes of rest between attempts.For each maximal attempt, subjects were asked to exert five seconds of maximal effort, while receiving standardized verbal motivation.Subjects completed each maximal attempt while seated on a chair in an erect position.The hip, knee, and elbow were flexed to a 90° angle and the shoulder was abducted and neutrally rotated.The forearm was in a neutral position and the wrist was slightly extended (0° to 30°).Subjects performed the test with a horizontal cylinder handgrip using the position II on the digital dynamometer, while the evaluator lightly supported this in its base.The best result (expressed in kg) of the three valid maximal attempts for each hand was chosen for statistical analysis.

Statistical Analyses
All values are reported as the mean ± standard deviation (SD).Normality and homoscedasticity assumptions for all data were checked with the Shapiro-Wilk and Levene's tests, respectively.When dependent variables had a non-parametric distribution, logarithmic procedures were conducted.Aside from age, all variables complied with the assumption of homoscedasticity (Leve-ne´s test).In the case of age, as the homogeneity of variances assumption is usually not as crucial as other assumptions (in particular in the case of balanced -equal n-designs) 39 , age-related analyses were performed with parametric tests and repeated using nonparametric methods to contrast.As no differences emerged, Pearson's r was used for all correlations.In addition, coefficient of determination (r 2 ) were employed in order to assess the best suited model (model quality) in relation with the age-related changes in MVIHS.For comparisons between dominant and non-dominant hand strength, the t test for dependent samples was used.For comparison between groups, a one-way ANOVA was performed with a post hoc Fisher LSD test to locate the pairwise differences between the mean values.The α level was set at p ≤ 0.05 for statistical significance.All statistical calculations were per-formed using the STATISTICA statistical package (Version 8.0; StatSoft Inc., Tulsa, OK, USA).For comparisons between frail and non-frail older adults, we considered the cut-points values of MVIHS stablished previously according to sex and body mass index 40 .

Results
High intra-class correlation coefficients were observed for MVIHS measurements (i.e., 0.92 and 0.95 for dominant and non-dominant hand, respectively), demonstrating high reliability of measurement.For males, 25 th , 50 th and 75 th percentile values of MIVHS were 28.3 kg, 32.9 kg, and 37.3 kg, respectively.For females, the values were 14.9 kg, 21.5 kg and 25.8 kg, respectively.The rest of the results are reported in Table 1, according to handedness, age and sex group.
In the whole sample of males, the coefficient of determination indicated a significant decline in MVIHS with increased age (-0.56 ± 0.07 kg per year; p < 0.0001), as wheel as in the whole sample of females (-0.50 ± 0.07 kg per year; p < 0.0001), with no significant difference between males and females (p = 0.52).
Comparisons between groups revealed that i) compared to the dominant hand, the non-dominant hand showed lower MVIHS in all groups, varying between -2.8 to -8.9% (Table 1); ii) compared to men, women exhibited lower dominant and non-dominant MVIHS in both age groups (Table 1); iii) compared to men and women ≥ 71 years old, men and women in the 65-70 age group showed greater dominant and non-dominant MVIHS compared to their respective sex group counterpart (Table 1).
The differences in MVIHS between genders and age groups is graphically depicted in figure 1. Significant differences (p < 0.05) in MVIHS were observed between all groups (Figure 1, upper portion).In addition, when MVIHS was expressed in relation to body mass (figure 1, lower portion), significant differences (p < 0.05) in MVIHS were also observed between all groups.Moreover, body mass predicted 30% of MVIHS values.When the body mass index was correlated with MVIHS, although a significant relationship was observed (p < 0,001), the coefficient of determination was rather low for both dominant (r 2 = 0.19) and non-dominant hands (r 2 = 0.17) in the whole group of older adults.
When older adults were divided into frail and non-frail groups according to their respective age and sex groups, as expected, for each of the four groups significant (p < 0.05) differences were observed between frail and non-frail older adults for MVIHS (Figure 2) when the values of both dominant and non-dominant hands were combined.Moreover, when the dominant and non-dominant MVIHS was compared in frail and non-frail older adults, greater (p < 0.05) values of strength were observed in the dominant hand (Figure 3).In addition, for a given hand (dominant; non-dominant), greater MVIHS values were observed in non-frail older adults as compared to their frail counterparts (Figure 3).

Discussion
The aim of this study was to describe MVIHS and its relationship with age, sex, and handedness in older adults.The main results indicated that MVIHS declined with age, is greater in men compared to women, and is greater in the dominant compared to the non-dominant hand.
The MVIHS may be associated with health   and functional status in older populations [14][15][16][17] and may be affected by several factors 5,6,41 .Our results indicate that age was one of these factors, as it was inversely associated with MVIHS, corroborating previous observations 22,42 .Similar to previous findings 21 , the decline in MVIHS was observed in both men and women.Our results also showed that the decline in MVIHS is similar in males and females with age.This result is comparable with previous reports  , suggesting that utmost care is needed in the oldest populations to reduce the probability of functional dependency due to reduced muscle strength.
In relation to dominance, our results indicated a greater MVIHS in the dominant hand compared to the non-dominant hand.This observation is relatively common 27,28 .More so, our results indicated a difference in MVIHS between the dominant and non-dominant hand of up to ~9% of the study cohort, comparable in magnitude (~10%) to values previously reported 27,28 .Although differences in MVIHS between dominant and non-dominant hands might depend on whether subjects are lefthand or right-hand dominant (i.e., right-handed subjects usually exhibit greater MVIHS in the dominant hand, while left-handed subjects show conflicting results) 24 , our sample was comprised mainly of right-handed subjects.Although a previous large-cohort study in Chile have described MIIHS values for Chilean older adults in their dominant hand 32 , our results expand previous findings, incorporating measures for both dominant and non-dominant hands.Such findings are of relevance, considering that during daily living older adults may use both hands to perform their activities.
In relation to sex, it is relatively common to observe that men exhibit greater MVIHS compared to women [9][10][11][12][14][15][16][17]19,20,46,47 , an observation corroborated by our results. Also, our resultsindicate that the difference between men and women remains constant up to an advanced age (i.e., ≥ 71 y), which had been previously observed in an older Chilean population 27 .Moreover, differences between men and women were observed in both dominant and non-dominant hands, meaning than sex-related differences in MVIHS might be common to contralateral limbs.In addition, our results indicated that differences between males and females remain even after adjusting for body mass (Figure 1).Further research is needed to corroborate if this can be extrapolated to lower limb muscles of older men and women.
Although few studies have been conducted in Chile focusing on the MVIHS of older adults and its relation with health 25,26 , the results from this study conducted in a southern region, are very similar to previous reports for older men and women from other regions of Chile 27,32 .This strengthens the external validity of previous observations and provides support for the recommendation of this measurement at a preventive level for Chilean older adults in the clinical practice at national level by the Chilean Ministry of Health 35 .
A potential limitation of the study is the lack of physiological measures (e.g., blood glucose, HOMA, plasma free fatty acids, blood pressure) and functional performance (e.g., up-and-go test).Future studies may aim to establish a relationship between cardio metabolic risk markers, functional performance and grip strength in Chilean older adult population.Moreover, the assessment of body composition, especially muscle mass, in future studies may help to better understand the relationship between MVIHS and the variables analyzed in the current study.Although the present work constitutes a starting point to establish normative values of MVIHS in Chilean older adults in the south of the country, more studies with larger samples are required.

Conclusion
In conclusion, MVIHS declined with age in older adults, with a similar rate in males and females, MVIHS is greater in men compared to women, and MVIHS is greater in the dominant compared to the non-dominant hand.

Figure 2 .
Figure 2. Maximal voluntary isometric handgrip strength (MVIHS) in frail and non-frail older adults according to sex and age.*denotes significant (p < 0.05) difference between frail and non-frail groups for a given sex-age group.

Figure 3 .
Figure 3. Maximal voluntary isometric handgrip strength (MVIHS) of frail and non-frail older adults according to handedness and age.*denotes significant (p < 0.05) difference between hands for a given age-group; #denotes significant difference between frail and non-frail groups for a given hand.