The combined deleterious effects of multiple sclerosis and ageing on neuromuscular function

.


Introduction
The demography of people with multiple sclerosis (pwMS) has changed markedly over the past decades, and now approximately onethird of all pwMS are 60 years or older (Mackenzie et al., 2014;Magyari and Sorensen, 2019;Wallin et al., 2019).Since many of the symptoms and signs of MS and ageing overlap, it is plausible that advanced age in pwMS will inflict combined or synergistic deleterious effects.
The neuromuscular function of the lower extremities appears particularly prone to such "ageing and MS" effects (Charlier et al., 2016;Clark and Manini, 2012;Thompson et al., 2013;Lomborg et al., 2022a;Sieljacks et al., 2020;Stagsted et al., 2021).Supporting this, our group has shown that knee extensor (KE) and plantar flexor (PF) maximal muscle strength (Fmax) are impaired to a greater extent in older pwMS compared to age-and gender-matched healthy controls (HC) (Sieljacks et al., 2020).From a clinical perspective, this is critical due to the association between lower extremity Fmax and several daily physical functional tasks (Ramari et al., 2020;Sieljacks et al., 2020;Lomborg et al., 2022b).
The rate of force development (RFD) is another measure of neuromuscular function that has received attention over the past years and represents the ability to rapidly increase muscle force during the start of maximal muscle contraction.RFD appears particularly critical in tasks requiring fast lower limb movement, such as when avoiding falling, not just in pwMS (Kristensen et al., 2023) but also in other populations undergoing neurodegeneration (Lomborg et al., 2022a;Pijnappels et al., 2008;Kamo et al., 2019).Interestingly, RFD has been found to be preferentially impaired compared to Fmax in young and middle-aged pwMS (Jørgensen et al., 2017;Scott et al., 2011;Taul-Madsen et al., 2022a) and during ageing (Thompson et al., 2013;Lomborg et al., 2022a;Varesco et al., 2019).However, no data presently exist on the combined or synergistic deleterious effects of MS and ageing on lower extremity RFD.
Altogether, a plethora of evidence supports the separate negative effects of ageing and MS on lower extremity neuromuscular function; however, the combined or synergistic effects have rarely been investigated.Therefore, the aim of the present study was to explore the effects of MS and ageing on 1) lower extremity Fmax and RFD, 2) lower extremity neuromuscular activation and muscle contractile properties, and 3) lower vs. upper extremity Fmax.We hypothesised that 1) lower extremity Fmax and preferential RFD would be impaired in pwMS vs. ageand sex-matched HC and that these impairments would increase with age, 2) lower extremity Fmax would be explained by reduced neuromuscular activation and altered muscle contractile properties, and 3) lower extremity Fmax would be impaired to a greater extent than upper extremity Fmax in pwMS.

Methods
The present cross-sectional study is a secondary analysis of data from Sieljacks et al. (Sieljacks et al., 2020), with further details provided in the main paper.This study was approved by the local ethics committee (Region Midtjylland,Denmark; and conducted in accordance with the Declaration of Helsinki.All enrolled participants provided written consent.

Participants
Recruitment of pwMS was achieved via the Danish MS Society and events at the local branches of the Danish MS Society.The inclusion criteria for pwMS were as follows: MS diagnosis, age ≥ 18 years, expanded disability status scale (EDSS) score ≤ 6.5, and the ability to independently attend testing.The exclusion criteria for pwMS were as follows: relapse ≤2 months prior to enrolment, medical comorbidities affecting the nervous or muscular system, pregnancy, having a pacemaker, blood pressure > 160/100 mmHg on the day of testing, and regular self-reported participation (≥ 3 sessions per week, ≥ 30 min sessions) in moderate-to-high intensity structured exercise ≤6 months prior to enrolment.HC were recruited via the network of participating pwMS and local senior centres.Inclusion and exclusion criteria for HC were identical to the pwMS criteria, with the exception of the MSspecific criteria.Both pwMS and HC were further sub-grouped into three age groups: young (age ≤ 44 years), middle-aged (age 45-59 years), and old (≥60 years).

Neuromuscular function 2.2.1. Dynamometer assessment
Muscle strength was assessed using an isokinetic dynamometer (Humac Norm, CSMi, Stoughton, MA, USA).The self-reported weaker leg performed maximal voluntary contraction (MVC) of both the KE and DF to determine the Fmax for both muscle groups.The torque was sampled at 1500 Hz using a TeleMyo Direct Transmission System and MyoResearch Software (Noraxon, Scottsdale, AZ, USA) and subsequently low-pass filtered (6 Hz).Torque data analysis was performed using custom-made software (MathWorks, MatLab 2020, Natick, MA, USA).
KE was tested in an upright seated position with a hip angle of 85 • and a knee angle of 70 • .For PF, participants were seated upright with their hip at 110 • , knee at 45 • , and ankle at 5 • plantar flexion.For a further description of the dynamometer setup, see Sieljacks et al. (Sieljacks et al., 2020).The testing order of KE and PF was randomised.Prior to testing both KE and PF, a standardised warm-up and familiarisation protocol was completed.The procedure consisted of two 5-s isometric contractions, one at 50 % MVC and one at 80 % MVC.The Fmax testing of the KE and PF consisted of 3-4 MVC trials with a 60 s rest period between trials.Each participant received strict instructions to contract as fast and forcefully as possible before all trials.Additionally, strong verbal encouragement and visual feedback of torque tracers on a computer screen were provided during all trials.
Hand grip strength was assessed using the Exacta™ Hydraulic Hand Dynamometer (North Coast Medical, Morgan Hill, CA, USA), which featured an adjustable grip mechanism to accommodate different hand sizes.The dynamometer was positioned between the medial phalanges and the base of the thumb.Three trials were conducted with the dominant hand, while the arm remained extended.
The Fmax was determined in the dynamometers for the KE, PF, and hand grip.Here, the largest force produced in the MVC trials was selected and expressed as torque normalised to body mass (Nm .kg − 1 ).From these MVC trials, RFD was determined from the onset of contraction to 50 ms (RFD 50ms ) and 200 ms (RFD 200ms ).Onset was defined as a torque increase of 2 standard deviations (SD) from the baseline torque level at rest.RFD was expressed as the change in torque normalised to body mass per time (Nm kg − 1 s − 1 ).RFD 50ms represented early phase RFD, and RFD 200ms represented late phase RFD (Varesco et al., 2019).Early phase RFD is generally attributed to neural properties (e.g., motor unit discharge rate), whereas late phase is generally attributed to peripheral factors (Folland et al., 2014;Maffiuletti et al., 2016;Andersen and Aagaard, 2006;Del Vecchio et al., 2019).

Interpolated twitch technique
ITT was used to determine neuromuscular activation, a proxy for neural drive, and muscle contractile properties represented by the resting twitch (RT).ITT was applied to both KE and PF, as previously described (Hvid et al., 2016).First, familiarisation with transcutaneous electric stimulation (direct current stimulator, model DS7A, Digitimer Ltd., UK) along with determining individual maximal stimulation current (i.e., systematic increase in current to no further increase in torque amplitude occurred) was performed.Electric stimulation was delivered by two 5 × 10 cm electrodes (Valutrode Lite, Axelgaard, Denmark) and was compromised by a doublet twitch with a 200 μs duration and 10 ms interstimulus delay.The electrodes were placed according to the SENIAM guidelines on the quadricep and gastrocnemius muscles.On the quadriceps muscle, the electrodes were positioned 15 cm proximal to the basis of the patella and 15 cm distal from the anterior superior iliac crest.On the gastrocnemius muscle, the electrodes were positioned on the most prominent bulge of the medial head.The ITT consisted of an MVC to which stimulation was applied at the force plateau and at rest immediately following the MVC.
T. Gaemelke et al.Neuromuscular activation was calculated as follows using torque at stimulation (T stim ), superimposed twitch peak (T ITT ), RT peak (T rest ), and MVC: Hence adjusting for trials in which the interpolated twitch stimulation was not delivered at the MVC peak.RT was expressed as peak torque normalised to body mass (Nm kg − 1 ).

Statistical analysis
Statistical analysis was performed using STATA 17 (STATA, IC 17, StataCorp, College Station, TX, USA).Outcome measures requiring normalisation (e.g., MVC) were normalised as described previously in the methods section.A mixed-effect linear regression was used with participant ID set as a random effect and age group (young, middle-aged, and old) and group (HC and pwMS) set as fixed effects.An age by group effect was investigated using the Wald test.A pairwise comparison of outcome margins was made across the factor variable levels used in the mixed-effect linear regression.Data are presented as the mean ± standard deviation (SD), and statistically significant differences were defined as p < 0.05.

Results
Across all groups, demographic and clinical characteristics were comparable between pwMS and HC, with the exceptions of a higher body mass, fat percentage, BMI, and a lower muscle percentage in pwMS (Table 1).In the young, a higher age, body mass, fat percentage, and BMI and a lower muscle percentage were observed in pwMS compared to HC (Table 1).

Neuromuscular function: Fmax and RFD
KE neuromuscular function was reduced in pwMS compared to HC for the collective samples of pwMS and HC.However, this reduction was preferential in RFD 50ms (RFD 50ms : − 48 % [− 60;− 36] vs. RFD 200ms : − 37 % [− 46;− 28]) vs. Fmax: − 24 % [− 32;− 17]) (Table 2).A corresponding pattern was observed across age groups, with more pronounced deficits in the old compared to the middle-aged and the middle-aged compared to the young (Fig. 1, Table 2).Furthermore, for all measures of KE neuromuscular function (RFD 50ms , RFD 200ms , and Fmax), an age × group interaction was observed, showing attenuated impairments with advanced age in pwMS compared to HC (Fig. 2 2).A corresponding pattern was observed for middle-aged and old pwMS, but with more pronounced deficits in the old vs. middle-aged groups (Fig. 1, Table 2).In the young, no differences were observed between pwMS and HC.Furthermore, for all measures of PF neuromuscular function (RFD 50ms , RFD 200ms , and Fmax), an age × group interaction was observed, showing attenuated impairments with advanced age in pwMS compared to HCs (Fig. 2, Table 2).Different trajectories were observed for pwMS, where decreases mainly appeared from young to middle age, and HC showed no changes.

Neuromuscular activation and muscle contractile properties
In the collective sample of pwMS and HC, KE neuromuscular activation was reduced in pwMS compared to HC (− 14 % [− 21;− 6]), whereas the KE resting twitch was comparable between pwMS and HC (3 % [− 10;16]).Across age groups, neuromuscular activation deficits were slightly more pronounced in the old compared to the middle-aged and the middle-aged compared to the young (Fig. 1, Table 2).Furthermore, for the KE resting twitch, an age effect was observed in pwMS, i.e., attenuated impairments with advanced age (Fig. 2, Table 2).
In the collective sample of pwMS and HC PF neuromuscular activation showed a trend towards lower levels in pwMS compared to HC (− 7 % [− 16;2]), whereas the PF resting twitch was reduced in pwMS compared to HC (− 29 % [− 41;− 17]).Across age groups, neuromuscular activation deficits were observed only in middle-aged pwMS, while resting twitch deficits were attenuated in the old compared to the middle-aged and the middle-aged compared to the young (Fig. 1, Table 2).Furthermore, for PF neuromuscular activation, an age × group interaction was observed, showing attenuated impairments with advanced age in pwMS compared to HC (Fig. 2, Table 2).For the PF resting twitch, we observed an age effect in pwMS, i.e., more pronounced impairments with advanced age (Fig. 2, Table 2).

Upper and lower extremity Fmax
Hand grip Fmax was reduced in pwMS compared to HC (− 12 % [− 19;− 4]) in the collective sample of pwMS and HC (Table 2).Across age groups, deficits in hand grip Fmax were observed in middle-aged (− 20 % [− 30;− 10]) but not in young (− 6 % [− 22;10]) or old pwMS All data are presented as mean ± SD except for participants and MS type which is presented as numbers.(− 6 % [− 19;6]) (Table 2).Furthermore, in both pwMS and HC an age effect was observed, i.e., attenuated impairments with advanced age (Table 2).Different trajectories were observed for pwMS, where decreases appeared mainly from young to middle age, and HC, where decreases appeared mainly from middle to old age.

Discussion
The present study explored the combined effect of MS and ageing on neuromuscular function, i.e., RFD and Fmax, as well as the underlying mechanisms responsible for Fmax, i.e., neuromuscular activation and muscle contractile properties, across the adult lifespan of pwMS and HC.We found that RFD 50ms , RFD 200ms , and Fmax were reduced between pwMS and HC within age groups, e.g., old pwMS compared to old HC (Fig. 1).Neuromuscular activation of pwMS showed similar decrements in KE compared to HC but with RT being unaffected; the opposite was observed for PF with decrements in RT, but neuromuscular activation was unaffected.Neuromuscular function diminished with increasing age (young > middle-aged > old) for both pwMS and HC.This was true for RFD 50ms , RFD 200ms , and Fmax for both KE and PF, but only pwMS presented reductions in muscle activation and RT.Interestingly, the trajectory of impairments with advanced age in RFD 50ms , RFD 200ms , and  Fmax differed between pwMS and HC (Fig. 2).The impairments in KE primarily occurred from young to middle age in pwMS and from middle to old age in HC.However, in PF, a continued impairment occurred from young to middle-aged to old in pwMS but not in HC.This suggests that the observed impairments in the neuromuscular function of MS and ageing are more pronounced in PF than in KE.Furthermore, the most substantial deficits in pwMS compared to HC were found in RFD.This was particularly evident in the case of RFD 50ms , which displayed deficits of up to 70 %.This observation underscores the compromise in neuromuscular function attributable to neurological factors (Folland et al., 2014;Maffiuletti et al., 2016;Andersen and Aagaard, 2006;Del Vecchio et al., 2019).
The different trajectories of KE and PF neuromuscular function in ageing, with slightly greater decrements in the latter compared to the former (Fig. 2), is an interesting finding.Speculatively, this may be caused by longer nerve tracts being at a higher risk of damage than shorter nerve tracts (Kurtzke, 2015), implying that neural degeneration is dependent on axonal length (Giovannoni et al., 2017).However, the neuromuscular function of the PF did not change with advanced age in HC.This may be explained by PF being a smaller muscle group that is loaded relatively more in older adults when engaging in daily activities, such as walking, compared to KE (Sieljacks et al., 2020;Kulmala et al., 2020).Hence, it appears that PF can be sufficiently stimulated in HC but not in pwMS to counteract the potential reduction in neuromuscular function seen during ageing.
RFD 50ms , RFD 200ms , neuromuscular activation, and RT have not previously been reported in older pwMS, either for KE or PF.However, RFD has previously been reported for KE but not PF in young and middle-aged pwMS (Kjølhede et al., 2015).Here, it was observed that RFD 50ms and RFD 200ms of the KE in young and middle-aged pwMS were lower than those observed in the present study (Kjølhede et al., 2015).This may be explained by the lower EDSS score in the present study population (approximately 1 EDSS point lower), indicating a lower disability status.In a study by Skurvydas et al., neuromuscular activation of KE was 63 % for young and middle-aged pwMS and 88 % for HC (Skurvydas et al., 2011).The HC enrolled in this study were comparable to the HC of the present study; however, their pwMS had lower neuromuscular activation than the pwMS of the present study (~66 % vs. ~72 %, respectively).Again, this difference may be explained by the lower EDSS scores of pwMS in the present study (approximately 1.5 EDSS lower).PF neuromuscular activation was comparable to previous findings for pwMS with a mean age of 55 years (Djajadikarta et al., 2020).Overall, this suggests that the comparators (young and middle-aged) for the old pwMSs in the present study were representative of neuromuscular function previously reported in the literature.
The utilisation of the ITT in the present study to determine neuromuscular activation and muscle contractile properties (i.e., RT) has not previously been done in older pwMS.In KE, the deficits in the neuromuscular function of old pwMS appeared to be more attributed to impairments in neuromuscular activation than RT (Fig. 1).In PF, however, deficits in the neuromuscular function of the old pwMS may be attributed to impairments in RT and, to a smaller extent, in neuromuscular activation.These data suggest that PF muscle characteristics differ between pwMS and HC during ageing and that neuromuscular activation seems to amplify this difference.These interpretations of neuromuscular activation and RT results should, however, be done cautiously, as they do not completely reflect the central and peripheral functions of the neuromuscular system (Gaemelke et al., 2021).Furthermore, the ITT has some methodological limitations (see: Methodological considerations).
Neuromuscular function is determined by neuromuscular activation, as discussed above, and muscle mass.Muscle mass, as well as muscle quality, has previously been found to be reduced in pwMS, as well as muscle quality (Ng et al., 2004;Wens et al., 2014).In the first paper published on the present data, the muscle mass percentage was included (Sieljacks et al., 2020).The impact of differences in muscle mass percentage was limited since muscle mass percentages were somewhat similar in pwMS and HC of an old age.This suggests that the declining trajectory of neuromuscular function in MS during ageing is more likely to be driven by a reduction in neural activation.However, a reduction in muscle mass may still play a role, as it does in the ageing process (Lexell et al., 1988;Kent-Braun et al., 2000).
Lower extremity Fmax was more compromised than upper extremity Fmax, as shown in Table 2, at least when relying on KE and PF compared to grip strength.In a previous review from our group, we observed similar results, with elbow flexion and extension being much less affected in pwMS compared to knee extension, knee flexion, plantar flexion, dorsi flexion, and hip flexion (Jørgensen et al., 2017).Although this difference may in part have been due to differences in lower vs. upper extremity physical activity behaviour (we assume that lower extremities become preferentially more sedentary/inactive), it does support the theory of length-dependent axonopathy that has previously been put forward (Kurtzke, 2015;Giovannoni et al., 2017).More studies are needed to further our understanding of this phenomenon.

Clinical implications
RFD has previously been shown to be preferentially sensitive in detecting differences across disabilities in association with EDSS (Taul-Madsen et al., 2022b).Such reductions in RFD seem particularly pronounced in old pwMS.In the present study, early-phase RFD was markedly reduced in the PF of older pwMS.This is important, as PF is crucially involved in walking, with walking acceleration originating partly from increased PF RFD and timing of torque onset (Wade et al., 2022).This could explain the walking deficits with increasing age in pwMS previously reported by Sieljacks et al. (Sieljacks et al., 2020) (of which the present study is a continuation) and Baird et al. (Baird et al., 2019).Furthermore, the reduced RFD of the KE could explain the diminished ability to recover balance when tripping/falling (Pijnappels et al., 2005), as 57 % of pwMS have been reported to fall within a 3 month period (Nilsagård et al., 2015).Therefore, increasing the RFD of the KE and PF in older pwMS could improve walking and reduce the risk of falling.RFD has previously been reported to increase in young and middle-aged pwMS following a 24-week progressive resistance exercise protocol by Kjølhede et al. (Kjølhede et al., 2015).However, a study evaluating a progressive resistance training intervention specifically targeting RFD in older pwMS is still pending.Considering that MS will inevitably lead to a substantial reduction of neuromuscular function, as evident from the findings of the present study, it seems crucial to initiate effective physical rehabilitation / exercise strategies in pwMS.

Methodological considerations
A few methodological considerations deserve mention.First, the present study utilised a cross-sectional design (i.e., by comparing pwMS and HC across different age groups), and this must be considered in relation to the interpretation of the results.Nothing can be inferred about longitudinal/prospective changes or causality.The comparison of fixed age groups across populations with different life expectancies may be problematic, as pwMS have a 7-year shorter life expectancy (Lunde et al., 2017).The exclusion of pwMS with EDSS ≥6.5 limits the generalisability of the findings to pwMS that have a preserved walking function.Second, the comparison of neuromuscular function between pwMS and HC with inherently different physical activity levels, as was the case in the present study (Sieljacks et al., 2020), complicates the interpretation of the results, as both reduced physical activity and disease affect neuromuscular function.Third, only KE and PF were evaluated in the present study, thus limiting the generalisability of the findings to these muscle groups.Other muscle groups may be affected differently, potentially under-or overestimating the true combined effect of ageing and MS on neuromuscular function based on the present study findings.Fourth, the widely used ITT applied in the present study presents methodological challenges regarding the interpretation of the results.Herein, the assessment of VA and its limitations have previously been discussed extensively (Haan et al., 2009;Shield and Zhou, 2004;Taylor, 2009), as well as what a meaningful difference or change in VA would be (Rozand et al., 2020).To reduce the methodological limitation and the influence it may have on the results, best practice guidelines were followed (Taylor, 2009).In the present study, ITT was applied to investigate neuromuscular function under maximal output.However, the results show that RFD is highly affected in pwMS throughout their lifespans, which could be due to an inability to quickly upregulate the central drive early during muscle contractions (Varesco et al., 2019).Further studies are needed to investigate this, applying techniques capable of measuring neural regulation, e.g., electromyography (Varesco et al., 2019).This may further elucidate the role of co-contractions in the neuromuscular function of pwMS, as shown to be present in walking task by an increased ankle stiffness (Massot et al., 2021).

Conclusion
The negative effects of MS and ageing combined in the neuromuscular function of PF and KE resulted in increased deficits in older pwMS compared to older HC.Despite larger reductions being found in PF than in KE, RFD was clearly the most diminished neuromuscular function (upwards of ~70 % deficits) in pwMS compared to HC across age groups.The neuromuscular function findings could partly be explained by reductions in VA and RT.These reductions in neuromuscular function appear most prominent in older pwMS, but reduction starts early and declines rapidly during middle age.

Declaration of competing interest
The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: Tobias Gaemelke reports was provided by Aarhus University Department of Public Health.If there are other authors, they declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Statistical differences are marked by the following: *: Ageeffect, a: Difference between HC and pwMS, b: Difference between young and middle-aged, c: Difference between middle-aged and old.Abbreviations: HC: Healthy control, MS: Multiple sclerosis, BMI: Body mass index, RR: Relapse reemitting, SP: Secondary progressive, PP: Primary progressive, EDSS: Expanded disability status score.

Fig. 1 .
Fig. 1.Percentage differences in neuromuscular function in knee extension and plantar flexion of people with MS (pwMS) compared to healthy controls (HC) across three age groups: young, middle-aged, and old.Data are presented as the mean with 95 % confidence intervals.Statistical differences are indicated by the following: a: different from HC, b: different from young, c: different from middle-aged, and *: age effect.Abbreviations: MS: Multiple sclerosis, HC: Healthy control, RFD: Rate of force development, Fmax: Maximal muscle strength, RT: Resting twitch, Y: Young, MA: Middle-age, O: Old.

Fig. 2 .
Fig. 2. Percentage differences in neuromuscular function in knee extension and plantar flexion between people with MS (pwMS) compared to healthy controls (HC) in the young age group.Data are presented as the mean and 95 % confidence Statistical differences indicated by the following: a: different from HC, b: different from young, c: different from middle-aged, and *: age effect.Abbreviations: MS: Multiple sclerosis, HC: Healthy control, RFD: Rate of force development, Fmax: Maximal muscle strength, RT: Resting twitch, Y: Young, MA: Middle-age, O: Old.

Table 1
Demographic and clinical characteristics.
All data are presented as mean ± SD.Statistical differences are marked by the following a: different from HC, b: different from young, c: different from middle-aged, *: age effect, and §: age*group effect.HC: Healthy control, MS: Multiple sclerosis, RFD: Rate of force development, Fmax: Maximal muscle strength, RT: Resting twitch.