Ageing is not associated with a decline in neuromuscular innervation or reduced specific force in men aged 20 and 50 years

Ageing is not associated with a declinein neuromuscular innervation or reduced speci®cforce in men aged 20 and 50 years

Jack Cannon, Kyle Tarpenning, Derek Kay and Frank E. Marino

Human Movement Studies Unit and Human Performance Laboratory, Charles Sturt University, Bathurst, Australia

Received 24 August 2000; accepted 29 December 2000

Correspondence: Kyle Tarpenning PhD, Human Movement Studies Unit, Charles Sturt University, Panorama Avenue, Bathurst NSW 2795, Australia

...........................................................................................................................................................................................................................................................................................................................

Summary

The exact mechanisms responsible for the decline in

strength with age are yet to be completely elucidated.

Three proposed mechanisms responsible for thedetrimental effect of increasing age on strength

include changes in muscle mass, speci®c force and/

or neuromuscular innervation. Thus, the purpose ofthis investigation was to determine if the age-related

reduction in peak isometric strength was primarily

associated with changes in muscle cross-sectionalarea, neuromuscular innervation and/or speci®c force.

The cross-sectional area of the knee extensor muscles(QCSA) was estimated in 13 younger men (YM;

20á8 � 1á6 years) and eight middle-aged men (MM;

53á8 � 4á2 years) prior to performing a series of fourmaximal voluntary isometric contractions on an

isokinetic dynamometer at an angle of 60° knee

¯exion. Peak force was determined and surfaceelectromyography was sampled from the rectus

femoris muscle during each maximal voluntary con-

traction. The cross-sectional area of the knee extensormuscles, peak force and integrated electromyography

(IEMG) were signi®cantly lower in the MM (P<0á01).

However, when peak force and peak IEMG valueswere corrected for QCSA, there were no signi®cant

differences between age groups. These results suggest

that the reduction in peak isometric force observed inthe MM was primarily associated with quantitative

changes in muscle mass, rather than reduced neuro-

muscular innervation or speci®c force. Therefore,

preserving muscle mass through resistance trainingmay signi®cantly reduce the age-associated differences

in peak strength and assist in promoting quality of life

and functional independence in older adults.

Keywords: cross-sectional area, muscle activation,

muscle atrophy, peak force, surface EMG.

Introduction

The deterioration in physical health with normal

ageing is associated with greater incidence of many

physiological disorders, such as glucose intolerance,osteoporosis and a general decline in physical capacity

which often results in a loss of mobility (Lindle et al.,1997). Consequently, normal ageing is associatedwith a transition from independent to dependent

living. Although a causal relationship is yet to be

established, numerous investigations report a corre-lation between the age-associated decline in physical

health and a reduction in muscular strength (Sinaki

et al., 1986; Bloesch et al., 1988; Hyatt et al., 1990).The results of such research suggest that the ability to

preserve muscular strength into old age may reduce

physiological disorders, maintain mobility and retainthe ability to perform activities necessary for daily

living. Preserving muscular strength into old age may,

therefore, assist in maintaining functional independ-ence and enhance the quality of life.

...........................................................................................................................................................................................................................................................................................................................

350 Ó 2001 Blackwell Science Ltd · Clinical Physiology 21, 3, 350±357

The age-related decline in muscular strength

becomes signi®cant during the sixth decade of life

and continues to decline with subsequent ageing(Larsson et al., 1978; Doherty et al., 1993; Hakkinen

et al., 1996). Much of the research attributes the

decline in strength with age to a reduction in musclemass resulting from a loss of type I and type II muscle

®bres, and the atrophy of those muscle ®bres

remaining (Lexell et al., 1988). However, a declinein muscle mass alone does not appear to fully account

for the reductions in strength with age (Roos et al.,1997). Insuf®cient voluntary muscle activation and/ora deterioration in muscle quality could result in a

reduction in strength independent of age-relatedmuscle atrophy.

It is well documented that ageing in humans is

associated with changes in the integrity of theneuromuscular system. Such changes include a

decrease in motor unit numbers (de Koning et al.,1988), an increase in motor unit size (StaÊlberg& Fawcett, 1982), an increase in the amplitude of

motor unit action potentials (StaÊlberg & Fawcett,

1982), reduced motor neurone conduction velocities(Mittal & Logmani, 1987) and a decline in the number

of a-motor neurones (Kawamurra et al., 1991).

Reduced neurotransmitter release and a reducedsensitivity of the muscle fascia to electrical stimulation

have also been observed in other species (Smith, 1992;

Robbins & Nakashiro, 1993). Therefore, it remainspossible that the decline in strength with age is also

related to changes in neuromuscular innervation.

Another possible mechanism for the decline instrength with age is reduced speci®c force. Existing

evidence suggests that older muscle may exhibit

reduced force per unit of cross-sectional area (CSA)than young muscle (Dowling et al., 1994; Jubrias et al.,1997). Bruce et al. (1989) observed a 20% decline in

force per muscle cross-sectional area in older subjects,and similar observations have been documented in

rodent muscle tissue (Brooks & Faulkner, 1991). The

mechanisms responsible for this phenomenon remainunclear, however, it has been proposed that qualitative

changes occur with in muscle ®bres, thereby affecting

force generation during cross-bridge formation(Brooks & Faulkner, 1991).

From previous literature, it appears that the most

probable mechanisms responsible for the decline instrength with age are a decrease in muscle mass, a

decline in neuromuscular innervation and/or reduced

speci®c force. Although the age-related decline in

strength has been quantitatively described for manyyears, no available study has attempted to determine

the effect of increasing age on muscle mass, neuro-

muscular innervation and speci®c force simulta-neously.

The purpose of this investigation was to determine

if the decline in peak isometric force with age wasprimarily related to quantitative changes in muscle

mass, a decline in neuromuscular innervation and/or

reduced speci®c force.

Methods

Subjects

Thirteen young men (YM) with a mean (�SD) age of

20á8 � 1á6 years, height 180á0 � 6á7 cm, mass 79á0 �8á8 kg and eight middle-aged men (MM) with a mean

(�SD) age of 53á9 � 4á2 years, height 178á0 � 6á5 cm,

mass 81á0 � 15á0 kg volunteered to participate in thestudy. Mean (�SD) BMI for the YM and MM was

24á4 � 2á2 and 25á3 � 3á3, respectively. All subjects

were non-smokers and considered healthy subsequentto completing the Physical Activity Readiness Ques-

tionnaire (PAR-Q) (Thomas et al., 1992). Subjects

were not involved in more than 2±3 h of moderatephysical activity per week and were classi®ed as

untrained. Written consent was gained from all

subjects prior to the commencement of the study.The investigation was conducted with the approval of

the Ethics in Human Research Committee of the

University.

Anthropometry

Anthropometrical data collectionPrior to testing, total thigh circumference (CT) wasdetermined on the preferred kicking limb (Overend

et al., 1992a) at the region of the thigh with the

widest girth (Knapik et al., 1996) using a measuringtape. Three circumference measurements were

recorded with the subject standing upright, with

minimal body weight on the preferred limb. Carewas taken to ensure that the measuring tape

remained perpendicular to the longitudinal axis of

J. Cannon et al. · Ageing, neuromuscular innervation and speci®c force............................................................................................................................................................................................................................................................................................................................

Clinical Physiology 21, 3, 350±357 · Ó 2001 Blackwell Science Ltd 351

the thigh during recording. Individual CT measure-

ments were averaged and the mean used for

subsequent calculations. All CT measurements wererecorded in millimetres. The coef®cient of variation

of CT measures in the YM and MM was 3á8 and

4á5%, respectively.

Calculation of knee extensor muscle

cross-sectional area

Total thigh cross-sectional area (TCSA) was deter-

mined anthropometrically according to Knapik et al.(1996). The anthropometrical model of Knapik et al.(1996) is based on the premise that the thigh iscomposed of concentric layers of skin, fat, muscle and

bone tissue. Therefore, an estimate of TCSA can be

obtained using CT measurements and equations forthe circumference, radius and area of a circle (Knapik

et al., 1996). Accordingly, TCSA for the YM and MM

were calculated using Equations (4)±(5), respectively,where RT is the total radius of the thigh,

CT � 2pRT �1�

RT � CT=�2p� �2�

TCSA � pR2T � p�CT=2p�2 �3�

Calculating the cross-sectional area for the kneeextensor muscle (QCSA) was based on the results

reported in Overend et al. (1992a). Using a computed

tomography technique, Overend et al. (1992a)observed that the knee extensor muscles occupied

38á41% of the TCSA in younger subjects (19±34 years;

BMI 23á1; n � 13), and 30á25% of the TCSA in oldersubjects (65±77 years; BMI 24á7; n � 11). Using these

percentages, an estimate of the QCSA for the subjects

that participated in the present investigation wasdetermined. The QCSA of the YM and MM was

estimated using Equations (3) and (4), respectively.

QCSAYM � TCSA � 0 � 3841 �4�

QCSAMM � TCSA � 0 � 3025 �5�

Mechanical assessment of muscle function

Peak force testingPeak force was assessed on the preferred kicking limb

(Overend et al., 1992a) using a Kin-Comä isokinetic

dynamometer (Chattanooga Group Inc., Hixon, TN,

USA). Subjects were secured to the dynamometer via

waist and shoulder straps. Because of the risk ofinterfering with the surface electrode used to collect

the electromyography (EMG) data, it was not poss-

ible to stabilize the active thigh during exercise. Theaxis of rotation for the dynamometer was visually

aligned with the lateral epicondyle of the femur with

the lower leg attached to the lever arm at the level ofthe lateral malleolus.

Once secured to the isokinetic dynamometer,

subjects performed a standardized warm-up andfamiliarization protocol prior to commencing the

peak testing. This warm-up consisted of six sub-maximal isometric knee extensions at an angle of

60° knee ¯exion, with the reference point of 0°being full extension. The ®rst two contractionswere performed at �50% of maximal effort, and the

second two contractions were performed at �70%

maximal effort. Finally, two further contractionswere performed at �90% maximal effort. A rest

period of 5 s separated each of the six submaximal

contractions.Subsequent to the warm-up, peak force testing

commenced. Peak force testing consisted of four

maximal voluntary isometric contractions of the kneeextensor muscles on the preferred leg. For peak force

testing, subjects were positioned in the Kin-Com

isokinetic dynamometer similar to that during thewarm-up. Peak isometric force of the knee extensor

muscle was performed at an angle of 60° knee ¯exion,

with the reference point of 0° being full extension. Arest period of 5 s separated each of the four maximal

voluntary contractions. Isokinetic dynamometer soft-

ware collected force data throughout each contractionat a rate of 100 Hz.

Analysis of peak force dataPeak force was determined as the single highest force

value produced over the four consecutive contrac-tions. Peak force was visually determined by Kin-

Com analysis software. From the anthropometrical

calculations of QCSA, the force generated per squarecentimetre of knee extensor muscle tissue (speci®c

force) was determined using Equation (6).

Specific force �Ncmÿ2�� Peak force �N�=QCSA �cm2� �6�

Ageing, neuromuscular innervation and speci®c force · J. Cannon et al............................................................................................................................................................................................................................................................................................................................


Electromyographical assessment

of muscle function

Application of the surface electrodePrior to the warm-up and peak force testing on the

isokinetic dynamometer, an EMG electrode with abandwidth of 50±450 Hz was attached to the belly of

the rectus femoris, midway along the anterior aspect

of the thigh. The reference electrode was placed onthe patella of the opposing limb. The skin under-

lying the electrode placement sites was carefully

prepared. Hair was shaved, and the outer layers ofepidermal cells were removed using abrasive paper.

Oil and dirt on the skin was removed by swabbing

with 70% isopropyl. A differential surface electrode(DelsysTM, Boston, MA, USA) was then placed on

the prepared muscle site linked to an EMG signal

acquisition apparatus (DelsysTM, Boston, MA, USA)and host computer (IBM ThinkpadTM, Armonk,

NY, USA). EMG data was sampled at rate of

1024 Hz for all tests, thus yielding raw signals.

Analysis of electromyographical dataRaw EMG data from the contraction where peak

force occurred was recti®ed and the movement

artefact removed using a high pass second orderButterworth ®lter with a cut off frequency of 15 Hz.

The data was subsequently smoothed using a low-

pass second order ®lter with a cut-off frequency of5 Hz, and integrated (IEMG). These procedures were

performed using MATLABTM gait analysis software

(The Mathworks Inc., Natick, MA, USA). The rawIEMG data was subsequently divided by the number

of data points to correct for differences in contractiontime for each subject. Finally, the raw IEMG data was

multiplied by 102á4 to correct for the ®lter and

sampling rate, which presents the data in units of volt-seconds (V s). Equation (7) describes the mathemat-

ical treatment of the raw IEMG data,

IEMG�V s�� Raw IEMG �V�=Data points� 102�4 �7�

Using the anthropometrical data, an estimate of the

quantity of myoelectrical activity (V s) generated per

square centimetre of knee extensor muscle (speci®cIEMG) was determined. For the YM and MM,

speci®c IEMG from the rectus femoris was deter-

mined using Equation (8),

Specific IEMG �V s cmÿ2�� IEMG �V s�=QCSA �cm2� �8�

Statistical analysis

Standard methods were used for the calculation of

descriptive statistics. Statistical analysis was per-

formed using SPSSä (Statistics for the SocialSciences, version 6á0) software. A one-way analysis

of variance (ANOVA) was used to identify between

group differences in TCSA, QCSA, peak force,speci®c force, peak IEMG and speci®c IEMG.

Correlations between QCSA, peak force and peak

IEMG were determined using a Pearson rank-ordercorrelation. The critical level for statistical signi®-

cance was set at P<0á05. Values are reported as the

mean � standard deviation (SD).

Results

Anthropometry

Thigh circumference measurements with the calcu-

lated TCSA and QCSA values for the YM and MM

are presented in Table 1. Statistical analysis of CT

measurements revealed no signi®cant difference

between the age groups (P � 0á1). The observed

mean (�SD) CT for the YM was 51á9 � 3á1 cm, andfor the MM 49á2 � 4á2 cm. No signi®cant differences

in TCSA were observed between the YM (215 �

26 cm2), and MM (194 � 33 cm2) (P � 0á12).Following the calculation of QCSA based on the

percentages reported in Overend et al. (1992a), a

signi®cant difference between the age groups wasobserved. QCSA for the YM was estimated as

83 � 10 cm2, and for MM 59 � 10 cm2 (P � 0á001).

Peak force

The peak force values produced by the YM weresigni®cantly higher compared with the MM

(P � 0á001). The YM were able to generate a peak

force of 632 � 116 N, whereas the MM generated418 � 85 N of peak force. The age-related difference



in peak force corresponded to a 34% reduction

between the ages of 20 and 50 years. A signi®cant

correlation between QCSA and peak force wasobserved in the YM (r � 0á83, P � 0á001) and MM

(r � 0á74, P � 0á03).

Speci®c force

Subsequent to correcting peak force values by QCSAno signi®cant difference was observed between

age groups (P � 0á35). The YM generated 7á6 �

0á9 N cm±2 compared with 7á2 � 1á1 N cm±2 gener-ated by the MM.

Integrated electromyography

When comparing both age groups in terms of

myoelectrical activity, the YM generated substantiallygreater quantities of myoelectrical activity compared

with the MM. The quantity of IEMG generated

during the maximal voluntary isometric contractionfor the YM was 14á6 � 5á5 V s, which was signi®-

cantly higher than the MM who generated

7á5 � 1á5 V s (P � 0á002). This difference corres-ponds to a 48% reduction in myoelectrical activity

between 20 and 50 years. The correlation between

QCSA and IEMG was not signi®cant in the YM(r � 0á31, P � 0á3) or MM (r � 0á51, P � 0á2).

Speci®c integrated electromyography

Once corrected for QCSA, it was observed that the

YM generated 0á18 � 0á1 V s of myoelectrical activ-

ity per square centimetre of knee extensor muscle.

This value was not signi®cantly different to the MM

who generated 0á13 � 0á1 V s of myoelectrical activ-ity per square centimetre of knee extensor muscle

(P � 0á1).

Discussion

Previous research suggests that the most probablemechanisms responsible for the reduction in strength

with age are: (i) a decline muscle mass associated with a

reduction in the total number of muscle ®bres and theatrophy of those ®bres remaining, (ii) a change in the

proportion of type I and type II muscle ®bres through

the selective loss of type II ®bres, (iii) a deterioration inneuromuscular innervation related to possible changes

in neurotransmitter release, reduced muscle sensitivity

to electrical stimulation, or the ability of the centraland/or peripheral nervous systems to adequately

stimulate older muscle tissue and/or (iv) qualitative

changes within muscle ®bres affecting cross-bridgeformation resulting in a subsequent decline in speci®c

force (Stanley & Taylor, 1993).

The purpose of the present investigation was todetermine if the decline in peak force with age was

primarily related to quantitative changes in muscle

mass, a decline in neuromuscular innervation, and/orreduced speci®c force. The results of the present

study indicate that the observed reduction in peak

isometric force between 20 and 50 years of age wasprimarily related to quantitative changes in muscle

mass, rather than changes in neural innervation or

speci®c force.

Table 1 Physical characteristics of subjects presented with peak and corrected force and IEMG values.

Group

CT

(cm)

TCSA

(cm2)

QCSA

(cm2)

Peak

force (N)

Speci®c

force (N cm±2)

Peak

IEMG (V s)

Speci®c

IEMG (V s cm±2)

Young men

Mean � SD 51á9 � 3á1 215 � 26 83 � 10a 632 � 116a 7á6 � 0á9 14á6 � 5á5b 0á18 � 0á1

Range 46á7±59á0 174±277 67±106 410±838 5á7±8á6 8á2±25á7 0á1±0á3

Middle-aged men

Mean � SD 49á2 � 4á2 194 � 33 59 � 10 418 � 85 7á2 � 1á1 7á5 � 1á5 0á13 � 0á1

Range 43á2±56á8 149±257 45±78 331±600 5á0±8á9 6á1±10á8 0á1±0á2

Values are mean � SD. CT is thigh circumference; TCSA is total thigh cross-sectional area; QCSA is the estimated knee-extensor muscle cross-

sectional area based on the results of Overend et al. (1992a). Statistical differences were determined by a one-way ANOVA.aIndicates signi®cant difference (P<0á001).bIndicates signi®cant difference (P<0á01).



An age-related decline in peak strength associated

with reduced speci®c myoelectrical activity would

suggest that the ability to innervate middle-agedmuscle is reduced, possibly as a result of changes in

motor neurones and/or a reduced central drive.

However, an age-related decline in peak strengthassociated with a reduction in speci®c force would

suggest that qualitative changes have occurred within

muscle ®bres, possibly as a result of changes in themechanics of cross-bridge cycling and/or changes in

®bre type proportions. In either case, such a decline

in strength with age would be less responsive toexercise intervention. The results of the present

investigation suggest otherwise.

Age-associated decline in muscle mass

The age-related decline in peak strength is clearly

associated with changes in muscle cross-sectional area

(CSA) (Narici et al., 1988; Overend et al., 1992b;Jubrias et al., 1997). Therefore, when investigating the

effect of ageing on neuromuscular performance it is

important to assess the relationship between peakforce and muscle CSA. Previous investigations study-

ing age-related changes in the composition of the

thigh region have commonly used young(19±38 years) and elderly (65±90 years) subjects (Rice

et al., 1990; Overend et al., 1992a). Although it is not

ideal to compare the thigh composition of the elderlysubjects (65±77 years) reported in Overend et al.(1992a) to the MM (50±59 years) in the present

investigation, it was the best alternative as similardata pertaining to middle-aged men (40±60 years) is

not available.

The comparable speci®c force values observed inthe present investigation suggests that the reduction

in peak strength with age was essentially related to the

quantitative changes in muscle mass, rather thanqualitative changes within the muscle. Previous

investigations studying the changes in speci®c force

with age are con¯icting. Jubrias et al. (1997) reportedthat signi®cant age-associated reductions in speci®c

force only became apparent when 23±65-year-old

subjects were compared with 65±80-year olds. How-ever, other investigations report that age-related

differences in speci®c force between 45 and 78 years

of age are not signi®cant (Frontera et al., 1991). Inaddition, Kent-Braun & Ng (1999) reported that

although a substantial reduction in peak force was

observed in older subjects compared with younger

subjects, no difference was apparent when correctedfor muscle CSA.

The discrepancy of previous results may re¯ect the

age range of subjects and the limitations of studydesigns. A longitudinal study of the changes in

speci®c force would contribute enormously in clar-

ifying these positions.

Age-related changes in ®bre type proportions

The most plausible mechanism responsible for a

change in ®bre type proportions with age is theselective loss of type II muscle ®bres, as a result of

selective denervation (Stanley & Taylor, 1993). As

muscle biopsy samples were not taken in the presentinvestigation, the current ®ndings are unable to

support or reject an age-related change in muscle

®bre type proportion. However, current researchprovides con¯icting evidence for the selective loss of

type II muscle ®bres. Larsson et al. (1978) reported an

increase in the relative proportion of type I muscle®bres in the vastus lateralis muscle between 26 and

62 years of age. Further evidence for the selective loss

of type II muscle ®bres with age is provided byKlitgaard et al. (1990). In contrast, Grimby et al.(1984) reported no signi®cant difference in the

proportion of type I and II muscle ®bres between66 and 100 years of age. Additionally, Lexell et al.(1986) reported similar results, where no signi®cant

difference was observed in the number type II ®bresin subjects ranging from 24 to 77 years old.

Age-associated decline in neuromuscular innervation

The insigni®cant difference in speci®c myoelectrical

activity with age observed in the present studyindicates that both age groups possessed a similar

capacity to innervate muscle tissue. This ®nding

suggests that between 20 and 50 years of age, nosigni®cant changes occur in neurotransmitter release,

muscle stimulation sensitivity, or the ability of the

central and/or peripheral nervous systems toadequately stimulate muscle tissue.

The present investigation is the ®rst study where

myoelectrical activity, or neural innervation, has beenanalysed per square unit of muscle tissue. Much of the



previous research on age-related changes in neuro-

muscular innervation has compared maximal volun-

tary strength values against strength values producedduring maximal voluntary efforts coupled with

electrical stimulation. Vandervoort & McComas

(1986) provide evidence that healthy men and womenaged 66±100 years are able to maximally activate their

ankle and dorsi ¯exors under isometric conditions as

supramaximal stimulation of the motor nerve failed toincrease maximal voluntary contraction strength.

Kent-Braun & Ng (1999) using the central activation

ratio method reported no signi®cant differencebetween maximal voluntary contractions and maximal

voluntary contraction with superimposed electricalstimulation in subjects aged 32 and 72 years.

Muscle quality

The results of the present investigation contradict the

notion of an age-related deterioration in musclequality resulting in a decline in speci®c force. As

discussed previously, reports of age-related changes in

speci®c force are con¯icting. The mechanism respon-sible for the age-associated decline in muscle quality

that leads to a reduction in speci®c force remains

unclear. Brooks & Faulkner (1991) studying themuscle of adult and aged mice suggest that either a

substance in the cytosol of ®bres in the muscle of aged

animals inhibits force development of the cross-bridges, or that intact ®bres are not fully activated

by calcium. In either case, formidable methodological

constraints must be overcome before the importanceof age-related reductions in speci®c force can be fully

determined in humans.

The major ®nding of the present study was thatdespite the signi®cant reduction in peak force and

peak IEMG with age, the differences were negligible

when corrected for QCSA. These results suggestthat the age-associated decline in strength up to

50 years is primarily related to quantitative changes

in muscle mass, rather than reduced neuromuscularinnervation or a deterioration in muscle quality.

Therefore, preserving muscle mass through

resistance training may signi®cantly reduce theage-related differences in peak strength and assist

in promoting quality of life and functional

independence in older adults.

Acknowledgements

This research was funded by a Charles Sturt UniversityPostgraduate Research Studentship. The authors sin-

cerely thank the subjects that participated in this study.

References

BLOESCH D., SCHULTZ Y., BREITENSTEIN E., JEQUIER E.& FELBER J. P. (1988) Thermogenic response to an oralglucose load in man: comparison between young andelderly subjects. J Am College Nutr, 7, 471±483.

BROOKS S. V. & FAULKNER J. A. (1991) Maximum andsustained power of extensor digitorum longus musclesfrom young, adult and old mice. J Gerontol, 46, B28±B33.

BRUCE S. A., NEWTON D. & WOLEDGE R. C. (1989)Effect of age on voluntary force and cross-sectional areaof human adductor pollicis muscle. Q J Exp Physiol, 74,359±362.

DE KONING P., WIENEKE G. H., VAN DER MOST VAN SPIJK

D., VAN HUFFELEN A. C., GISPEN W. H. & JENNEKENS

F. G. I. (1988) Estimation of the number of motor unitsbased on macro-EMG. J Neurol Neurosug Psychiatry, 51,403±411.

DOHERTY T. J., VANDERVOORT A. A. & BROWN W. F.(1993) Effects of ageing on the motor unit: a brief review.Can J Appl Physiol, 18, 331±358.

DOWLING J. J., KONERT E., LJUCOVIC P. & ANDREWS D. M.(1994) Are humans able to voluntarily elicit maximummuscle force? Neurosci Lett, 179, 25±28.

FRONTERA W. R., HUGHES V. A., LUTZ K. J. & EVANS W. J.(1991) A cross-sectional study of muscle strength andmass in 45±78 year old men and women. J Appl Physiol,71, 644±650.

GRIMBY G., ANLANSSON A., ZETTERBERG C. & SALTIN B.(1984) Is there a change in the relative muscle ®brecomposition with age? Clin Physiol, 4, 189±194.

HAÊ KKINEN K., KRAEMER W. J., KALLINEN M., LINNAMO V.,PASTINEN U. M. & NEWTON R. U. (1996) Bilateral andunilateral neuromuscular function and muscle cross-sectional area in middle-aged and elderly men andwomen. J Gerontol: Biol Sci, 51A, B21±B29.

HYATT R. H., WHITELAW M. N., BHAT A., SCOTT S. &MAXWELL J. D. (1990) Association of muscle strengthwith functional status of elderly people. Age Ageing, 19,330±336.

JUBRIAS S. A., ODDERSON I. R., ESSELMAN P. C. & CONELY

K. E. (1997) Decline in isokinetic force with age: musclecross-sectional area and speci®c force. P¯ugers Arch-Eur JPhysiol, 434, 246±253.

KAWAMURRA Y., OKAZAKI H., O'BRIEN P. C. & DYCK P. J.(1991) Lumbar motoneurons in man. I. Number and



diameter histogram of alpha and gamma axons of ventralroots. J Neuropathol Exp Neurol, 36, 853±860.

KENT-BRAUN J. A. & NG A. V. (1999) Speci®c strength andvoluntary muscle activation in young and elderly womenand men. J Appl Physiol, 87, 22±29.

KLITGAARD H., MANTONI M., SCHIAFFINO S. et al. (1990)Function, morphology and protein expression of ageingskeletal muscle: a cross-sectional study of elderly menwith different training backgrounds. Acta Physiol Scand,140, 41±54.

KNAPIK J. J., STAAB J. S. & HARMAN E. A. (1996) Validity ofan anthropometric estimate of thigh muscle cross--sectional area. Med Sci Sports Exerc, 28, 1523±1530.

LARSSON L., SJODIN B. & KARLSSON J. (1978) Histochemicaland biochemical changes in human skeletal muscle withage in sedentary males, age 22±65. Acta Physiol Scand, 103,31±39.

LEXELL J., DOWNHAM D. & SJOSTROM M. (1986) Distri-bution of different ®bre types in human skeletal muscles:®bre type arrangement in m. vastus lateralis from 3groups of healthy men between 15 and 83 years. J NeurolSci, 72, 211±222.

LEXELL J., TAYLOR C. C. & SJOSTROM M. (1988) What isthe cause of the ageing atrophy? Total number, size andproportion of different ®bre types studied in whole vastuslateralis muscle from 15 to 83 year old men. J Neurol Sci,84, 275±294.

LINDLE R. S., METTER E. J., LYNCH N. A. et al. (1997) Ageand gender comparisons of muscle strength in 654women and men aged 20±93 yr. J Appl Physiol, 83,1581±1587.

MITTAL K. R. & LOGMANI F. H. (1987) Age-relatedreductions in 8th cervical ventral nerve root myelinated®bre diameters and numbers in man. J Gerontol, 42, 8±10.

NARICI M. V., ROI G. S. & LANDONI L. (1988) Force ofknee extensor and ¯exor muscles and cross-sectional areadetermined by nuclear magnetic resonance imaging. EurJ Appl Physiol, 57, 39±44.

OVEREND T. J., CUNNINGHAM D. A., PATERSON D. H.& LEFCOE M. S. (1992a) Thigh composition in young andelderly men determined by computed tomography. ClinPhysiol, 12, 629±640.

OVEREND T. J., CUNNINGHAM D. A., KRAMER J. F., LEFCOE

M. S. & PATERSON D. H. (1992b) Knee extensor and knee¯exor strength: cross-sectional area ratios in young andelderly men. J Geront: Med Sci, 47, M204±M210.

RICE C. L., CUNNINGHAM D. A., PATERSON D. H. & LEF-

COE M. S. (1990) A comparison of anthropometry withcomputed tomography in limbs of young and aged men.J Geront: Med Sci, 45, M175±M179.

ROBBINS N. & NAKASHIRO S. (1993) Connectivity amongplasticity, regeneration and aging at the neuromuscularjunction. Adv in Neurol, 59, 47±52.

ROOS M., RICE C. L. & VANDERVOORT A. A. (1997) Age-related changes in motor unit function. Muscle Nerve, 20,679±690.

SINAKI M. J., MCPHEE C., HODGSON S. F., MERRITT J. M.& OFFORD K. P. (1986) Relationship between bonemineral density of spine and strength of the back exten-sors in healthy postmenopausal women. Mayo Clinic Proc,61, 415±419.

SMITH D. O. (1992) Celluar and molecular correlates ofaging in the nervous system. Exp Gerontol, 23, 399±412.

STAÊ LBERG E. & FAWCETT P. R. W. (1982) Macro EMG inhealthy subjects of different ages. J Neurol NeurosurgPsychiatry, 45, 870±878.

STANLEY S. N. & TAYLOR N. A. S. (1993) Isokinetic musclemechanics in four groups of women of increasing age. EurJ Appl Physiol, 66, 178±184.

THOMAS S., REDDING R. & SHEPARD R. J. (1992) Revision ofthe Physical Activity Readiness Questionnaire (PAR-Q).Can J Sp Sc, 17, 338±345.

VANDERVOORT A. A. & MCCOMAS A. J. (1986) Contractilechanges in opposing muscles of the human ankle jointwith aging. J Appl Physiol, 61, 361±367.



Documents

Ageing is not associated with a decline in neuromuscular innervation or reduced specific force in men aged 20 and 50 years