METHODS

Participants

Four men (mean (SD) age 20.5 (0.74) years, height 177.3 (3.5) cm, mass 74.3 (3.3) kg) and four women (mean (SD) age 21.1 (0.84) years, height 167.3 (3.4) cm, mass 63.5 (5.5) kg) volunteered to participate in the study, which was approved by the school’s ethics committee. All participants gave written informed consent and had not been involved in a weight training programme within the preceding six months.

Experimental design

After a one repetition maximum test on both arms to estimate the dynamic concentric strength performance capability of the elbow flexor musculature, this musculature in the non-dominant arm was trained at 60% of one repetition maximum, three times a week for two weeks, increasing to 70% of maximum voluntary strength performance for a further two weeks. Three days later, both arms were subjected to an eccentric exercise protocol designed to stress the elbow flexors.

Concentric training protocol

Concentric training occurred three times a week for four weeks. Participants were required to stand with the elbow fully extended and lift a dumbbell from full elbow extension up to full elbow flexion with the non-dominant arm. The non-dominant arm was selected to maximise the effect of the concentric training programme. It was felt that the dominant arm could be more influenced potentially by concentric and eccentric activities associated with habitual daily physical activities and intrusions associated with the physiological incompatibility of endurance and strength exercises. To ensure that no eccentric training occurred, the dumbbell was removed from the participant at the end of the concentric contraction (full elbow flexion). Each individual performed three sets of 10 repetitions at 60% of one repetition maximum during weeks 1 and 2. This was increased to three sets of 12 repetitions at 70% of one repetition maximum during weeks 3 and 4. Each set was separated by a three minute rest period. There followed a three day rest period to minimise possible symptoms of EIMD.

Eccentric exercise protocol

Three days after performing the concentric training protocol (during which time the possible symptoms of EIMD would be reduced), each participant performed five sets of ten maximal isokinetic eccentric contractions on each arm on a Kin Com isokinetic dynamometer (500H; Chattecx, Chattanooga, Tennessee, USA) to induce muscle damage. After each eccentric contraction, from full flexion to full extension at 60°/s, the investigator returned the arm to elbow flexion so that no concentric exercise occurred. A one minute rest separated each set of 10 contractions and participants were encouraged to exert maximal effort throughout the exercise bouts.

Measurements

Measurements of isometric strength, upper arm circumference, relaxed arm angle (RAA), and muscle soreness were recorded for both the concentrically trained and untrained arm before the concentric training programme, before an eccentric exercise protocol, one hour after exercise and at 24, 48, and 72 hours.

One repetition maximum

At the onset of the study all participants performed a one repetition maximum test on both arms to provide an estimate of dynamic concentric strength performance capability of the elbow flexor musculature that could be used subsequently to regulate the exercise intensity of the concentric training protocol. The participant was asked to stand with the elbow fully extended and lift a dumbbell from full elbow extension to full elbow flexion. The weight of the dumbbell was increased by 2.5 kg until the participant was no longer able to lift the weight. The latter was recorded as the subject’s one repetition maximum. A three minute rest separated each repetition in order to minimise fatigue.

Isometric strength

Isometric strength was assessed using a Kin Com isokinetic dynamometer. The participant was placed on a portable treatment couch (Darley, Lostwithiel, Cornwall, UK) in the supine position. A restraining strap was placed around the body, in line with the point at which the elbow flexed, to prevent any extraneous movement. The elbow was flexed to anatomical 80°, which was identified with a goniometer. A semipermanent marker was used to mark the axis of rotation at the elbow, and points on the humerus and forearm, which were proximal and distal to the axis of rotation. These marks were used to relocate the goniometer position in order to set the anatomical 80° joint angle on subsequent occasions.

After a warm up of three submaximal and two maximal isometric contractions, participants performed two three second maximal isometric contractions separated by a one minute rest. A visual display unit, which displayed force in real time, was used to encourage maximal efforts.³⁵ The mean of the two three second isometric contractions was taken to be representative of the force production capability. Lower coefficients of variation and higher intraclass correlations have been reported for mean torque.³⁶ Strength was reported as a percentage of baseline values in order to remove any pretest differences.

Upper arm circumference

Arm circumference was measured with an anthropometric tape measure. Both arms were marked with a semipermanent pen at the mid-belly of the biceps brachii. Provided that the scores lay within a 5 mm range, the mean value of two measures was recorded.

Relaxed arm angle

RAA was assessed with the participant in the standing position with the arm in a relaxed anatomical position by their side, similar to that described by Cleak and Eston.³⁷ A goniometer (Baseline; Physiomed, Chester, UK) was used to measure RAA. The RAA at the elbow was subtracted from 180° so that changes in the elbow angle could be expressed relative to the anatomical zero position. In this way, a reduction in resting elbow angle was recorded as a positive change from zero. As described above, marks were placed on the forearm, humerus, and elbow for accurate relocation of these points on subsequent occasions.

Soreness

Soreness was evaluated using a visual analogue scale, which ranged from 0 (no soreness) to 10 (worst soreness ever). Participants rated the level of soreness on the concentrically trained and untrained arm during unassisted active elbow flexion and extension on each arm.

Design analysis

Data were analysed using separate two factor (arm by time) analyses of variance with repeated measures on both factors. Alpha was set at 0.05. The sphericity assumption was tested using Mauchly’s test of sphericity. In the event of any violation of this assumption, Greenhouse-Geisser (_GG) corrections were applied. Significant results were followed up using an adapted Tukey’s post hoc analysis for repeated measures.³⁸

RESULTS

Strength

Concentric training of the non-dominant arm resulted in a 14.4 (7.2)% increase (p<0.01) in isometric strength, compared with a non-significant change of 1.8 (8.0)% in the contralateral arm (arm by time F(1,7) = 12.0, p<0.05). Figure 1 shows the absolute values.

After the eccentric exercise bout there was a greater overall decrease in strength in the concentrically trained arm (F(1,7) = 19.54, p<0.01). A significant arm by time interaction (F(4,28) = 6.2, p<0.01) showed that strength in the concentrically trained arm was significantly lower than baseline at 24, 48, and 72 hours, whereas strength in the untrained arm was significantly lower than baseline at one hour only (fig 2).

Arm circumference

There was no significant change in arm circumference by the end of the four week training programme (F(1,7) = 0.9, p>0.05). After the eccentric exercise bout, there was a significant time by arm interaction on arm circumference (F(4,28) = 3.4, p<0.05). The circumference of the concentrically trained arm was significantly greater than baseline at 48 and 72 hours after eccentric exercise. The concentrically trained arm also had a significantly greater circumference than the control arm 72 hours after eccentric exercise. There were no other differences (fig 3).

Relaxed arm angle

The concentric training programme resulted in a significant reduction in RAA in the concentrically trained arm at the end of the four week training programme, with no changes in RAA of the untrained arm (arm × time F(1,7) = 3.4, p<0.01) (fig 4).

After eccentric exercise, there was a significant reduction in the RAA across time in both arms (F(4,28) = 17.1, p = 0.00), with the highest values being recorded at 24 hours. Although both conditions illustrated a similar pattern of decrease in RAA, it is notable that the untrained arm had a higher RAA before damage, which remained significantly higher at all times (F(1,7) = 15.2, p<0.01). Signs of recovery were evident in both arms 48 hours after eccentric exercise; however, neither arm had recovered by 72 hours (fig 5).

Soreness

After the eccentric exercise bout, soreness increased significantly in both the control and the concentrically trained arm (F(1.8,12.5) = 10.05, p<0.01_GG), with greater soreness being observed in the concentrically trained arm (F(1,7) = 17.1, p<0.04). A significant arm by time interaction (F(4,28) = 2.8, p<0.05) showed that soreness was greater in the concentrically trained arm at 24, 48, and 72 hours. Soreness in the untrained arm was also significantly greater than baseline at 24 and 48 hours with values returning to baseline by 72 hours. However, soreness in the concentrically trained arm was still significantly greater than baseline 72 hours after eccentric exercise (fig 6).

DISCUSSION

The four week concentric training programme resulted in a 15% increase in isometric strength compared with a 1% increase in the untrained arm. No significant changes were observed for arm circumference by week four, which suggests that the increase in strength may have occurred as the result of a neural adaptation.^39,40 RAA and associated elbow flexion in the trained arm increased significantly from baseline to the end of the four week concentric training programme (15.0 (6.9)° v 22.9 (7.2)° respectively). Changes in the untrained arm (14.5 (6.5)° v 16.7 (7.4)° respectively) were not significant and were within 95% confidence limits associated with measurement error.

The mechanical properties of the sarcomere and its surrounding layers of elasticity can be represented by “three-element” models focusing on contractile, series elastic, and parallel elastic elements.⁴¹ The length-force relation of the musculotendinous unit indicates a potential for certain passive structures within muscle, such as connective tissue sheaths—for example, endomysium, perimysium, epimysium, aponeurosis, tendon—and cytoskeletal elements—for example, intermediate filaments, titin, nebulin—to exert a force when unstimulated muscle is stretched, for example by the action of gravity on the mass of the forearm during the assessment of RAA. Any adaptation resulting from the four week concentric training programme, in which there is remodelling and hypertrophy within the connective and cytoskeletal elements and associated changes to the mechanical properties of these parallel elastic elements including greater stiffness, may account for at least some of the reduction in RAA. This would be especially so because the passive mechanical contribution would be expected to be greatest at the longest muscle lengths.⁴²

The increase in maximal voluntary isometric strength following the concentric training programme suggests that, even though it was short in duration and undertaken at moderate levels of intensity (60–70% of isometric strength), it was still sufficient to induce performance improvements despite being unlikely to have activated fully the motor unit pool. It is probable that neural adaptation was the principal mechanism by which increased performance was affected, given that the cross sectional area of muscle was unchanged.^43,44 Various mechanisms would be implicated, including concurrent changes to recruitment and discharge rate⁴⁵ at a potentially decreased muscle length (discussed below),⁴⁶ potentiation of reflex responses,⁴⁷ and increased excitability of the alpha-motoneurone pool.^48,49 However, such mechanisms of neural adaptation that may underpin the observed changes in strength performance should not have influenced RAA to any great extent given the static and passive nature of the test.

The decrease in RAA may indicate a degree of muscle shortening, which may be attributable to a decrease in sarcomeres in series, which theoretically could decrease the length of the muscle. A decrease in sarcomere number would also theoretically increase muscle stiffness. Whitehead et al²⁸ reported an increase in passive torque after a short concentric training programme, which is consistent with this theory. Transmission of force from the muscle to the bone is facilitated by muscle stiffness,^30,31 as this reduces the amount of stretch in the non-contractile component of the muscle as the contractile component shortens and therefore transfers the contractile force to the bone sooner.³² It is therefore conceivable that concentric exercise may shorten a muscle, making it stiffer, thereby enhancing the efficacy with which force can be translated from the muscle to the bone.

After eccentric exercise, symptoms of EIMD were observed in both arms. However, more pronounced symptoms were observed in the concentrically trained arm, which concurs with other findings.^27,28 Isometric strength was significantly lower in the concentrically trained arm at 24, 48, and 72 hours after the eccentric exercise bout, whereas isometric strength in the untrained arm had returned to the baseline value by 24 hours.

Arm circumference was significantly greater in the concentrically trained arm at 72 hours, compared with baseline levels in both arms and in comparison with the untrained arm at 72 hours. This is attributed to a greater inflammatory response involving movement of fluid, plasma proteins, and leucocytes into the damaged tissues,⁵⁰ and is in accordance with previous studies.^37,51–53 The results also concur with those of Whitehead et al,²⁸ who observed greater swelling (leg volume) in the concentrically trained leg than in the untrained leg after a bout of eccentric exercise.

Soreness was also greater in the concentrically trained arm at 24, 48, and 72 hours. These results differ from those of Whitehead et al,²⁸ who reported no difference in soreness between the concentrically trained and untrained ankle plantar flexors after eccentric exercise. The difference in results is most likely due to differences in protocol of the two studies. The RAA decreased significantly in both arms after damage, but was significantly lower at every time point after damage in the concentrically trained arm.

The decrease in RAA in the concentrically trained arm may provide indirect evidence of a reduction in sarcomeres, which would be commensurate with a decrease in muscle compliance after intensive concentric training. This predisposes the muscle to more severe symptoms of muscle damage after eccentric exercise. Passive torque has been used to evaluate muscle stiffness and is positively associated with symptoms of EIMD.^28,29 A possible explanation may be that, because of muscle shortening, more sarcomeres have to perform at lengths that correspond to the plateau or descending section of the length-tension curve, where overextension and greater disruption of sarcomeres in series is more likely.^14–16 It has been suggested that sarcomeres in a stiffer musculotendinous unit are longer at any given point in the muscle contraction,^30,31 and it is well documented that EIMD is more severe after eccentric exercise at long muscle length.^12,14–16 It is likely therefore that more sarcomeres are overextended in a stiffer musculotendinous unit, which allows less scope for the tendon and series elastic component of the muscle to stretch.²⁹

The plasticity characteristics of muscle are exemplified by the change in sarcomere number after immobilisation of cat soleus muscle in a short or long muscle position.^54,55 The reduction in the number of sarcomeres in series that occurs with immobilisation in a shortened position is associated with an increase in passive muscle tension and a shift in the length-tension relation to the left.^56,57 If concentric training induces a reduction in the number of sarcomeres, then differences in the length-tension relation of muscle contraction would also be expected. In rats, eccentric training has been associated with a greater number of sarcomeres than concentric training.^25,26 Rats that followed an eccentric exercise programme attained peak torque at longer muscle lengths than rats that followed a concentric exercise programme.^25,26

An alternative explanation for the greater severity of symptoms of EIMD in the trained muscle could be that the increases in strength induced by the concentric training programme resulted in greater eccentric forces during the eccentric exercise bout. This could result in more force per fibre during the eccentric loading. Ploutz-Snyder et al²⁷ attributed the greater damage and decrease in strength to a greater eccentric loading potential induced by the increase in concentric strength. Although it is beyond the scope of this study to estimate the force per fibre during the exercise bout, it is unlikely that this is the case. Previous research has shown that concentric training has a minimal effect on eccentric increases in strength.^58,59 In addition, it is notable that, although the concentrically trained, non-dominant arm increased significantly in strength (p<0.01), there was no significant difference between the two arms before or after the concentric strength training protocol. Respective absolute values for the trained and control arms before the strength training programme were 240.9 (62.3) N and 258.6 (75.9) N, and those after the training programme were 275.5 (60.4) N and 262.6 (79.2) N. The significant interaction of arm by time on strength was due to a greater relative increase in strength in the trained arm compared with the untrained arm. It is therefore difficult to conclude from this study that the more pronounced symptoms of EIMD were attributable to a greater loading potential in the concentrically trained arm.

The deployment of concentric muscle actions may be appropriate clinically during key phases of physical rehabilitation in order to provide maximum protection for avascular surgically reconstructed tissue while facilitating prevention of fibrosis and muscular atrophy. It may be that in such circumstances the intensity of either rehabilitation or functional activities would be insufficient to induce substantive muscle damage routinely. However, in situations such as intermediate phases of rehabilitation involving accelerated progress towards functionally relevant activities, where the clinician or exercise scientist may be presented with a choice as to the most appropriate intervention strategy, the findings from this study offer several important considerations for optimised rehabilitation practice.

The concentric training regimen used in this study improved this performance capacity by about 15% compared with baseline and the untrained control limb. It is interesting to note that the time course of strength restoration toward baseline levels after the bout of eccentric exercise showed appreciably different patterns for the concentrically trained and untrained limbs. For example, one hour after the bout of eccentric exercise, the net loss of strength performance, taking account of the gains from the concentric training (about 15% compared with the initial baseline) and effects of the eccentric loading (about 24% loss compared with performance after training), was about 9%. This loss of performance compares favourably with a limb that has not undergone training in which the net loss of performance was about 17% (fig 2). In the context of rehabilitation, this point in time after a session involving eccentric muscle actions may be associated with routine daily living and workplace related activities with inherent risks to musculoskeletal integrity. As such, the net effects on strength performance of concentric training before eccentric stresses may still confer biologically significant advantages of increased protection from injury compared with no training. Furthermore, concentric training may facilitate musculoskeletal health by additional means associated with the potentiation of reflex responses, increased excitability of the alpha-motoneurone pool, and a mechanically stiffer system.⁶⁰ The results suggest that net changes to strength performance may be an important consideration for rehabilitation for up to 48 hours after the bout of eccentric exercise. During this time, the strength performance of the concentrically trained arm was at least the equivalent of the untrained arm.

In conclusion, this study extends the understanding of the scope of the effects of prior concentric training in increasing the severity of EIMD to that involving an upper limb exercise model. It provides further evidence, in addition to that presented by Whitehead et al²⁸ and Ploutz-Snyder et al,²⁷ that the inclusion of concentric conditioning in rehabilitation programmes will tend to exacerbate the severity of EIMD in subsequent unaccustomed exercise. However, where concentric conditioning is indicated clinically, the net effect of conditioning outcome and EIMD may still confer an enhanced strength performance and capability to dynamically stabilise a joint system.