*2.5. Statistical Analysis*

For the study group, demographic data were reported as the arithmetic mean ± standard deviation or with absolute and relative frequencies. The normal distribution was tested with Shapiro–Wilk's test.

The collected time graphs during measurements were analysed for errors. The peak torque was considered missing in case errors were found. Other missing data were due to fatigue, muscle pain or discomfort of the participant.

Clinical measurements were compared between the moments of measurements (at 2, 4 and 6 h) and between the active versus static behaviour with one-way ANOVA for repeated measures, separately for each flexion, foot and degree. The *p*-value for the comparisons between the moments (different measurements in time) and the *p*-value for the comparisons between active and static behaviour was reported. The number of data entered in the oneway ANOVA with repeated measures was 8 for each moment (one measurement for each participant).

When all the data were analysed (576 measurements), multivariate analysis was performed with linear mixed models for repeated data because the data were not independent between the participants (each participant had 12 measurements in each moment). Independent factors were considered, and they were entered into the analysis as fixed factors: the active/static behaviour, the moments, the flexion, the foot and the degree. The data were analysed and reported to the first considered moment at 2 h. The data at 0 h were not available due to the protocol design.

Arithmetic means of the clinical measurements were computed. These arithmetic means were compared between the moments of the measurements with *t*-test for paired samples for normally distributed data and with Wilcoxon signed-rank test for non-normally distributed data. Post-hoc analysis was performed using Bonferroni correction.

Correlations between two parameters were analysed by computing the Pearson and Spearman coefficients of correlation.

The *p*-value was considered statistically significant for values smaller than 0.05. Analysis was performed using SPSS application (manufactured by IBM Corp., Armonk, NY, USA, 2017) [58].

#### **3. Results**

#### *3.1. Participant's Characteristics*

Our participant's characteristics are described in Table 1. The average participants' age was 35.88 years old, with a minimum age of 23 and a maximum age of 58 years; three participants (37.5%) were male; four participants (50%) reported an average daily number of steps higher than 6000; five participants (62.5%) had a normal body mass index (BMI). For the studied sample, the HGS, CC, CRST, CRT, chair raise test and muscle pain and fatigue moment of onset during CRST, CRT and chair raise test for all participants and results are reported in Table 1.


**Table 1.** Demographic and anthropometric data of the sample.

<sup>1</sup> Total sample (*n* = 8); <sup>2</sup> BMI—Body Mass Index; <sup>3</sup> AHGS—Average Hand Grip Strength; <sup>4</sup> ACC—Average Calf Circumference; <sup>5</sup> CRST—Calf Raise Senior Test; <sup>6</sup> ACRT— Average Calf Raise Test; <sup>7</sup> no.—number.

#### *3.2. Clinical Measurements Results*

In Table 2, the descriptive statistic parameters for peak ankle torque during MVIC (dMVIC) in the case of active and static behaviour for each moment in time were presented and analysed with an ANOVA repeated measure for each foot, flexion and degree. The number of data entered in the ANOVA repeated measure analysis was *n* = 8 (one measurement of dMVIC for each participant).

When we analysed all the data with linear mixed models without taking into consideration the foot, the flexion or the degree, we found that the active versus static behaviour (*p* = 0.005) and the moments (2 h vs. 6 h *p* = 0.040, 2 h vs. 4 h *p* = 0.128) had a significant effect on the dMVIC. The number of data entered in the linear mixed models' analysis was *n* = 288 (for each subject 36 measurements of dMVIC).

The averages of dMVIC measured on different flexion, foot and degree on different moments per static/active behaviour are presented in Figure 8. There were significant differences between dMVIC after two hours compared with dMVIC after six hours (*p* = 0.019). There were no significant differences between dMVIC after two hours compared with dMVIC after four hours (*p* = 0.224) and also between dMVIC after four hours compared with dMVIC after six hours (*p* = 0.815). The average dMVIC in the case of active behaviour was significant statistically greater than the average dMVIC during static behaviour.

**Figure 8.** Impact of the moments of testing and the behaviour on dMVIC.


**Table 2.** Descriptive statistics (arithmetic mean +/− standard deviation; median (25th; 75th percentile)) for dMVIC in the cases of active and static behaviours.

\* dMVIC—peak torque (maximum value of maximal voluntary isometric contraction with pedal off-set correction), \*\* *p*-value from ANOVA with repeated measure.

When we tested the other factors: age, gender, BMI, average steps, AHGS, ACC, CRST, CRT, chair raise test, muscle pain during CRST, muscle pain during CRT, muscle fatigue during CRST and muscle fatigue during CRT in a repeated-measure ANOVA model, we did not find any significant influence on average dMVIC.

We analysed the impact of the daily activity as the average number of steps (Table 3) or as the difference between the group of routinely active and routinely sedentary, but we did not find any significant statistical correlation (r = −0.313, *p* = 0.450) or association (*p* = 0.882) with dMVIC.

**Parameters Pearson/Spearman Coefficient of Correlation** *<sup>p</sup>* Age (years) 0.229 0.586 BMI <sup>1</sup> (kg/m2) 0.580 0.132 Foot length (cm) 0.483 0.226 Average no. <sup>6</sup> of daily steps (steps/day) −0.313 \* 0.450 AHGS <sup>2</sup> (kg) 0.573 0.137 ACC <sup>3</sup> (cm) 0.359 \* 0.382 CRST <sup>4</sup> (no. <sup>6</sup> of repetitions) 0.535 0.171 Muscle pain during CRST <sup>4</sup> (no. <sup>6</sup> of repetitions) 0.569 0.141 Muscle fatigue during CRST <sup>4</sup> (no. <sup>6</sup> of repetitions) 0.365 0.373 ACRT <sup>5</sup> (no. <sup>6</sup> of repetitions) 0.411 0.360 Chair raise test bilateral 0.048 0.911 Muscle pain ACRT <sup>5</sup> (no. <sup>6</sup> of repetitions) −0.252 \* 0.585 Muscle fatigue ACRT <sup>5</sup> (no. <sup>6</sup> of repetitions) −0.131 0.779

**Table 3.** Correlation between average dMVIC and the other parameters.

<sup>1</sup> BMI—Body Mass Index; <sup>2</sup> AHGS—Average Hand Grip Strength; <sup>3</sup> ACC—Average Calf Circumference; <sup>4</sup> CRST—Calf Raise Senior Test; <sup>5</sup> CRT—Calf Raise Test; <sup>6</sup> no.—number. \* Spearman coefficient of correlation.

When we tested the correlation with The Pearson and Spearman coefficients of correlation, the same not-significant statistical relationship between average dMVIC and the other factors was found (Table 3).

#### **4. Discussion**

Based on previous studies, we hypothesised that a sitting posture could negatively influence foot and ankle muscle strength.

One day of inactivity was associated with an unhealthy potential on human health [28], and sitting showed secondary negative effects even in the active population [29].

In our study, both routinely active and routinely sedentary participants showed a significant statistically decrement in average peak ankle torque when subjected to six hours of prolonged sitting. In our group, foot and ankle muscle strength suffered a reduction over time even when participants were subjected to six hours of low to moderate physical activity.

Chronic unloading of negative effects on the Achilles' tendon was demonstrated, but when resistive exercises were added, some preventive effects were observed [34].

Considering the results demonstrated by Reeves et al. [34], we could estimate that when some type of exercise is associated, preventive effects of inactivity might be installed.

In our study, we demonstrated that all eight participants, when subjected to low to moderate physical activity, showed higher values for average peak ankle torque when compared to the values obtained when subjected to sedentary behaviour.

In the case of traumatic events, immobilisation effects have been reported. A decrease in ankle plantar flexor muscle strength has been observed after one week of immobilisation, but no significant differences were seen after two days of ankle immobilisation. The accumulated effects of immobilisation in time could explain the differences [33].

In our study, we evaluated the acute effects of six hours of inactivity on the plantar flexors' muscles through the measurement of ankle torque in nontraumatic events. There were significant differences between dMVIC after two hours compared with dMVIC after six hours (*p* = 0.019). We should further consider that the same study could be replicated related to traumatic events.

In our study, only after six hours of prolonged sitting a significant decrement in average peak torque was observed, with no significant differences between measurements at two hours and at four hours. We can state that in our group of participants placed under static behaviour, ankle torque decreased after the accumulated effects of prolonged sitting.

In patients suffering from diabetes, some electrophysiological reports analysed the soleus muscle activity in different sedentary postures, comparing chair-sitting with squatlike sitting [32].

In our group of participants, when sitting activity (static behaviour) was compared with a low to moderate physical activity (active behaviour), we found that even in healthy individuals', inactivity has negative potential on foot and ankle muscles strength.

Optimisation of the effects of muscle activity for promoting health in the general population and reducing sedentary behaviour strategies, daily contraction duration of skeletal muscle and the role of contractile duration were studied [7].

Our study reports the effects on ankle torque when individuals were placed in a shorttime sedentary behaviour. Due to the insignificant reduction in torque at four hours but the significant decrement in torque after six hours, new considerations should be evaluated for proper preventive strategies to reduce the negative effects of prolonged sitting on foot and ankle muscle strength.

Based on our findings on the acute effects of six hours of sedentary behaviour on ankle torque, we might consider that while individuals adopt a prolonged sitting position, interposing activity-based breaks after at least four hours of inactivity might positively influence lower limb muscle performance. We only found a significant statistically decrement in ankle torque after six hours of inactivity, but no significant statistically decrement was found after two hours and four hours. These findings could help in establishing more rigorous prevention strategies for reducing sedentary behaviour effects of prolonged sitting in individuals. Such prevention strategies could address modifications of break time and break frequencies, time of work in sitting postures and routine behaviour modifications, as other previous studies showed [59].

In sports medicine and rehabilitation medicine, foot and ankle muscle strength assessments are essential, and one indicator of muscle performance tested in clinical practice is measuring ankle torque by assessing MVIC [60].

As no agreement has been stated on the most appropriate method for measuring the strength of the foot's intrinsic muscles [38], our study analysed peak ankle torque by measuring MVIC using a reliable [41] and reproducible [42] custom-made electronic dynamometer.

As per our knowledge, no particular study has assessed the impact of sedentary behaviour on ankle torque using a custom-made electronic dynamometer; therefore, describing the measurement system in detail as an innovative way to measure muscle strength in relation to inactivity was of great importance.

The hand grip strength cut-off values and CC have been stated in relation to muscle strength reduction in sarcopenia [50,53], and some correlations with lower limb muscle strength have been reported [51].

We found no significant statistically relationship between average dMVIC and the average values of HGS and CC, probably due to the small sample (8 participants).

The number of repetitions during the Calf Raise Senior Test (CRST) [54,55] has previously been used to assess plantar flexor muscle strength.

We applied CRST to our group of participants, and muscle pain/muscle fatigue appearance and the onset moment of both fatigue and pain were registered. We found no significant statistical relationship between average dMVIC and the average values of CRST fatigue and pain during CRST.

The Calf Raise Test (CRT) for evaluating calf muscle properties is a well-known clinical method [56], and based on a systematic review, the standards have been reported.

When we tested CRT in our group, muscle pain and muscle fatigue appeared, and the onset moment of both fatigue and pain were registered. We found no significant statistical relationship between average dMVIC and the average values of CRT, fatigue and pain during CRT.

The chair raise test, another commonly used method for assessing lower limb muscle parameters [57], was performed on our group of participants, and the total number of repetitions of chair raises in one minute was registered, with no significant statistical relationship between average dMVIC and the average values of the number of repetitions during the chair raise test.

Although a massive increment in sedentary-spent time was reported [16], with increased exposure to risks in all demographics and age groups [17], no particular study has analysed the impact of short-time sedentary time on ankle torque.

We found no significant statistical relationship between average dMVIC and the other factors, such as age, BMI, foot length and average no. of daily steps.

We analysed the impact of the type of daily activity (defined by the average number of steps, which differentiates the two groups of routinely active and routinely sedentary participants), but we did not find any significant statistical correlation (r = −0.313, *p* = 0.450) association (*p* = 0.882) with dMVIC.

Despite sitting being associated with physical inactivity and further health risks, the amount of sitting time linked with risks for human health has not yet been defined [36,37].

Understanding the acute effects of prolonged sitting on ankle torque could better frame the long-term effects of sedentary behaviour on both routinely active and routinely sedentary individuals.

Correlations between physical inactivity and diseases need a better understanding.

In this study, we took into consideration only the measurements at two hours from the initiation of the behaviour, but not measurements at the baseline. The authors think that they ensured that after maintaining two hours of static/active behaviour, all participants had the same level of physical activity before the initiation of measurements. However, a better approach can be considered. Ideally, the study should have implemented a baseline measure in the protocol with a levelled physical activity for each participant. We considered our study limitations to have not established a levelled baseline of physical activity for the participants before being subjected to the measurement in both types of behaviour. To consider the baseline measurements when reporting the impact of the testing moments and the impact of two types of behaviour on ankle torque would have been desirable.

We took into consideration only the measurements at two hours from the initiation of the behaviour. This further ensured that after maintaining two hours of static/active behaviour, all participants had the same level state of physical activity before the measurements.

This particular limitation is also due to the inconsistency of the participant's data recovered from their wearable devices (smartwatches). We could only recover an average of daily steps from the participant's last month of activity. Unfortunately, the data extracted from such devices does not represent the exact type of undertaken activity, with physical activity nor physical fitness being identifiable from the recovered data. To better identify the exact type of activity through the data derived from smartwatches, a more systematic control should be considered and eventually correlated with participants' reports of activity questionnaires/scales [61].

Another limitation of our study is due to the small sample. Because of that, the comparison between samples with eight data (Table 2) was not found to be statistically significantly different, but the same data, when compared with multivariate techniques, were statistically significant.

Further studies should implement the baseline measurement in the protocol when a levelled physical activity before the measurements are ensured and identify the exact type of physical activity through modern technology or by using validated questionnaires.

#### **5. Conclusions**

The more time participants maintained either short-term static or short-term active behaviour, the lower the average peak ankle torque resulted in both situations. Both routinely active and routinely sedentary participants showed a decrement of force in time when maintaining both types of behaviours, with the sitting position being associated with a lower value of average peak ankle torque during maximal voluntary isometric contraction.

Future studies should target the establishment of a threshold for the time spent in a sitting position, sedentary behaviour, in relation to foot muscle strength and establish whether breaking the routine during a specific activity might positively change the muscle force results.

Our force measurement results could complete ergonomic improvements for the achievement of healthy foot status in individuals spending prolonged time in a sitting position and especially when sitting while working.

Future studies should consider repeating the experiment in other types of groups of participants and possibly in groups affected by different conditions.

**Author Contributions:** Conceptualisation, I.I.D. and F.G.P.; methodology, T.P. and F.A.; software, T.P.; validation, S.B. and F.A.; formal analysis, C.I.B. and T.P.; investigation, I.I.D.; resources, E.-A.P.; data curation, T.P. and F.G.P.; writing—original draft preparation, I.I.D.; writing—review and editing, F.G.P.; visualisation, I.I.D., F.G.P., T.P. and F.A.; supervision, M.I., F.L.B. and N.D.R.; project administration, I.I.D. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Institutional Review Board Statement:** The study was conducted in accordance with the Declaration of Helsinki, and approved by Ethics Committee of University of Medicine and Pharmacy "Victor Babes" Timisoara, released and registered under Nr. 50/21.09-14.10.2020.

**Informed Consent Statement:** Informed consent was obtained from all subjects involved in the study. Written informed consent has been obtained from the patient(s) to publish this paper.

**Acknowledgments:** We thank all participants that voluntarily enrolled in the study.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**

