Effects of Repetitive Handgrip Training on Endurance, Specificity, and Cross-Education.

Key Words: Cross-training, Fatigue, Handgrip, Low-frequency, Muscle fatigue, Resistance training.

Training specificity refers to improvements in a task that are a unique function of the variables chosen for an exercise training program.[1] With few exceptions,[2] training specificity has received widespread support in the literature.[1,3] Contraction type,[4,5] training task,[6] degree of maximal effort,[6] velocity of movement,[7] joint position,[4,8] and time of day[9] are all factors that can influence training specificity.

Most studies examining training specificity have involved primarily high-resistance, low-repetition exercise using 75% to 100% of maximal effort.[1,2,3,10,11] Few investigations have examined the effects of exercising with low percentages of maximal effort (30%) on training specificity.[12]

The improvement in maximal force or endurance (work) of the contralateral limb induced by exercising the ipsilateral limb is called "cross-education."[3,12-14] Much of what we know about cross-education comes from studies examining changes in the ability to produce maximal force (for a review, see Morrissey et all) and is frequently attributed to neurological adaptations (facilitate synergists and inhibit antagonists) learned during training and then unconsciously applied to the untrained limb.[3,15] Few investigations, however,, have assessed whether endurance training with a low resistance (30% of maximal voluntary isometric contraction [MVIC]) enhances endurance in the contralateral limb.[12]

Yasuda and Miyamura[12] concluded that increased blood flow from ipsilateral rhythmic handgrip training (30% of MVIC) for 6 weeks induced a systemic blood flow adaptation that was the primary cause of improved rhythmic handgrip endurance in the contralateral untrained limb. To better understand the effect of learning on cross-education, we extended the work of Yasuda and Miyamura.[12] The subjects in our study trained with a similar protocol,[12] but we also had a group of subjects who performed the handgripping task with a load that would not be expected to produce a physiological adaptation. Under these conditions, if there was a similar training effect in this low physiological load training group compared with the untrained limb of a group that trained to fatigue with a 30% load, we believed it would suggest that adaptations (vascular) usually associated with exercise intensity are not primary contributors to cross-education.[12] Accordingly, the purpose of this study was to compare the effect of rhythmic right handgrip training (6 weeks) on bilateral rhythmic handgrip work (RHW), bilateral isometric handgrip endurance (IHE), and bilateral MVIC.

Method

Subjects

Twenty-four male volunteers were randomly assigned to a regular training group (n = 8), a low-level training group (n = 8), or a control group (n = 8). The age, height, and weight for each group are presented in Table 1. There was no difference ([F.sub.2,22] = 0.932, P [is greater than] .05) among the 3 groups for age, height, and weight. None of the subjects were competitive athletes who regularly trained their upper extremities, and none of the subjects had a history of cervical spine or upper-extremity impairments. Subjects were asked to refrain from racket sports and sports that involved throwing during the study. All subjects were right-hand dominant, as determined by the hand used to sign their name. Each subject gave written informed consent. Each subject was paid $10 per training session in an effort to ensure adherence to training. No subjects from any of the 3 groups missed any of the sessions.

Table 1.
Descriptive Statistics for Subjects in the Regular Training Group,
the Low-Level Training Group, and the Control Group

Group                 X      SD       Range

Regular training
  Age (y)            25      5.3      21-27
  Height (m)          1.8    0.09   1.71-1.89
  Weight (kg)        75      5.6    6.94-78.3

Low-level training
  Age (y)            27.2    4.8      23-30
  Height (m)          1.69   0.12   1.65-1.74
  Weight (kg)        76.2    6.2    68.1-77.9

Control
  Age (y)            26.2    7.1      22-29
  Height (m)          1.74   0.11   1.72-1.79
  Weight (kg)        74.5    5.3    68.6-79.4

Measurements

Three to 5 days prior to training and the day following 6 weeks of training, the following measurements were obtained bilaterally on all subjects: the mean of 2 maximal isometric handgrip force (MVIC) measurements, maximal RHW (in joules) at 30% of the dominant right hand's MVIC, and the time (IHE) (in minutes) that an isometric handgrip could be held using 30% of the dominant hand's MVIC. The intraclass correlation between the 2 MVICs was .98, supporting the high reproducibility for these repeated measurements. Thirty percent of the MVIC (mean of 2 measurements) was used for the endurance testing and training because previous research has shown that rhythmic handgrip training at this percentage produces an enhancement in peak blood flow[12,16] and endurance[12] and a decrease in muscle sympathetic nerve activity.[17] These findings suggest that endurance (rhythmic handgrip work) is enhanced with this protocol.

Instrumentation

A handgrip ergometer was specially designed so that rhythmic handgrip work could be measured with the subjects in the supine position. This ergometer was similar to the dynamometer used by Yasuda and Miyamura.[12] Briefly, the ergometer consisted of a 1.67-cm metal dowel (gripping dowel) attached to an adjustable cable that, in turn, was draped over a low-friction pulley(*) secured to the end of a table. The cable ran parallel with the table before traversing the pulley, forming a 90-degree angle over the end of the table. The cable ran perpendicular to the floor, where weights were attached. A ratchet clutch connected the pulley to a counter so that the exact excursion of the load during lifting could be measured. One complete pulley revolution corresponded to lifting the load 0.1413 m. The counter odometer could detect to 0.1 revolution; thus, fractions of revolutions were measured. A metronome delivered an audible sound every second to control the rate that the load was lifted (positive work) and returned (negative work) to the floor. An upright dowel (stabilization dowel) was firmly secured to the table (perpendicular) and positioned in the palm of the hand between the thumb and index finger during the rhythmic handgripping to prevent the hand from slipping toward the load. This stabilization method ensured a reproducible excursion of the load throughout the testing and training program.

Isometric handgrip force was measured using a load cell (Genisco AWU-250([dagger])) that was placed in series with 2 plates incorporated into a separate handgripping apparatus. The linearity, hysteresis, repeatability, and accuracy were all less than 2% of full scale with routine calibrations between each test. The load cell signal was monitored via an oscilloscope.

Experimental Protocol

Testing. All testing was done 3 to 5 days before and 1 day after 6 weeks of training. The subjects were tested in the supine position, with both shoulders abducted to 90 degrees. We used 6 weeks of exercise because previous studies[12,16] showed an effect of endurance training in 4 to 6 weeks using this workload. First, we determined the MVIC bilaterally. The proximal phalanges of the digits formed a 45-degree angle with the metacarpal bones during all isometric handgrip tests. This angle was considered the mid-range for the subjects' grip aperture. The subjects held the maximal contraction for 3 seconds for each of 2 trials. One minute separated the MVICs.

Next, we determined the maximal work capacity for rhythmic handgripping using a load equal to 30% of the dominant hand's MVIC. We determined the distance the load moved by placing the gripping dowel over the distal crease of the long and ring fingers (distal interphalangeal joint) with the fingers extended. Upon finger flexion, the counter measured the distance the load moved. Subjects were encouraged to complete their full range of finger flexion for each handgrip repetition. Each subject lifted and gradually returned the load to the floor over this predetermined excursion once every 2 seconds. During this test, the subject received no visual feedback. The subject raised the weight over the first second and lowered the weight over the next second. The metronome kept the cadence. The exercise continued until the subject could not maintain the cadence, as indicated by missing 2 consecutive positive work phases. The easily identifiable endpoint of this endurance test is why other researchers have used this protocol extensively.[12,16] Subjects were asked to minimize other muscle contractions from the contralateral arm during testing. Random tests of the left or right arm occurred on separate days.

On a separate day, the subjects held 30% of the dominant arm's MVIC until the force fell 50% of its initial value to establish the IHE. This measure would provide an indication of the static handgrip endurance between the pretraining and posttraining conditions. The subjects received feedback regarding the force level from the oscilloscope. This test used the same mid-range grip position.

Training. All training was conducted with the subjects in the same supine position as that described for the testing procedures. No subject received training bouts of MVIC or IHE tests. The regular training group exercised their right hand by repetitively gripping and releasing a load equal to 30% of MVIC, until the stopping criteria for the testing procedure were met or their training time progressed to over 2 hours. In the first week of training, the average time for training was 22 minutes. As the training progressed, only 2 subjects achieved the 2-hour training time limit, which occurred in their fifth week of training. Therefore, during the final 2 weeks, these 2 individuals trained for over 2 hours. The post-training test, however, stopped when the subjects were not able to continue following 2 consecutive incomplete contractions. The subjects received a 5-minute rest at the conclusion of each day's first training bout, followed by a second training bout. The second bout was always considerably shorter than the first bout due to the fatigue developed during the first test. The subjects repeated the training sessions 5 days per week for a total of 6 consecutive weeks.

Subjects in the regular training group also used the handgrip ergometer with their left hand (untrained) either before or after (random) their right-hand training session, using a near-zero load (0.15 kg) and the same 1 second on/1 second off rate. The low-level training group also used the handgrip ergometer bilaterally, using the same near-zero load and rate. This method served to maximize the contributions of learning and familiarity with the testing apparatus in the low-level training group and in the untrained limb of the subjects in the regular training group. The control group did not receive any training. On average, the near-zero load was equivalent to 0.005% of MVIC. This low-level training was comparable to opening and closing a hand at a 1 second on/1 second off rate. Thus, the low-level training group really represented a type of control group that used the training apparatus 5 times a week, similar to the training group; Because we did not expect the low-level training group to fatigue at these small loads, we had them perform the average number of repetitions used by the right hand for the regular training group from the previous day. The low-level training group used this low-level training 5 days per week for 6 weeks, similar to the number of visits for the regular training group.

Throughout the training period, subjects were reminded to avoid recreational exercises involving the upper extremities that exceeded their normal activities before beginning the study. Following the 30 sessions, all subjects were retested bilaterally for MVIC, RHW, and IHE. The training and testing procedure for RHW was identical (except for weeks 5 and 6 for 2 subjects, as previously described), so changes were available on a daily basis for this variable. All subjects were extremely motivated, did not miss any sessions, and tolerated the exercising well. Some soreness was reported within the first 2 weeks, but the major discomfort was primarily at the end of each bout of the fatiguing protocol.

Data Analysis

Because each complete pulley revolution corresponded to lifting the load 0.1413 m, the total work (RHW) was calculated by multiplying the 30% load force (LF) (in newtons) by the exact number of pulley revolutions (PR) to the nearest tenth of a revolution, times the pulley conversion factor (0.1413 m), times the 2 work phases (positive and negative). Thus, the equation used to calculate the RHW (in joules) was as follows[15]:

RHW = (LF) (PR) (0.1413 m)(2)

Comparisons were made before and after training for each of the 3 dependent variables (RHW, MVIC, IHE) partitioned by group and by hand. Interaction and main effects were assessed using a 2-factor repeated-measures analysis of variance (ANOVA). Significance was set at the .05 level. When interactions were significant, a simple effects analysis was completed. We determined mean weekly values of RHW for the regular training group's fight hand for descriptive purposes. In addition, linear regression analysis was used to describe the change in RHW over the 6-week period.

Using the sample variability as an estimate, a .05 chance for a Type I error, a sample size of 8, and a 50% change in the dependent variables after training, we had over 80% power in this study.

Results

The pretraining values for all dependent variables were not different between the regular training group and the low-level training group or between the right and left hands (Tab. 2). The right-hand RHW of both the regular training group and the low-level training group increased, but at different magnitudes. The right-hand RHW after repetitive submaximal training (30%) increased from 1,192J to 15,396J (P [is less than or equal to] .05) (Tab. 2). The average number of repetitions increased from 83.8 to 832 following the 6 weeks of training. The low-level training group also showed an increase from 1,129J to 1,473J (P [is less than or equal to] .05) in fight-hand RHW just by completing 6 weeks of low-level training (Tab. 2) (average number of repetitions increased from 75.6 to 93.4). The posttraining right-hand RHW of the regular training group (30% of MVIC), however, was greater than that of the low-level training group and the control group (15,396J versus 1,473J and 1,195J, respectively) (P [is less than or equal to] .05). This finding supports the notion that the 30% maximal load used for the regular training group induced a greater ability to perform work than the low-level training.

Table 2. Average ([+ or -] SE) Rhythmic Handgrip Work (RHW), Maximal Isometric Handgrip Force (MVIC), and Isometric Handgrip Endurance (IHE) Before (Pretest) and After (Posttest) Training for Regular Training, Low-Level Training, and Control Groups

                       Right Hand
Group                  Pretest                Posttest

RHW (J)
  Low-level training   1,129 [+ or -] 122      1,473 [+ or -] 98(a)
  Regular training     1,192 [+ or -] 181     15,396 [+ or -]
                                               5,504(a,b)
  Control              1,138 [+ or -] 135      1,195 [+ or -] 127

MVIC (N)
  Low-level training   447.5 [+ or -] 33.59    460.3 [+ or -] 33.23
  Regular training     424.1 [+ or -] 19.69    434.5 [+ or -] 18.14
  Control              428.2 [+ or -] 21.74    432.6 [+ or -] 17.61

IHE (rain)
  Low-level training   4.288 [+ or -] 0.391    3.947 [+ or -] 0.324
  Regular training     4.435 [+ or -] 0.286    4.122 [+ or -] 0.438
  Control              4.160 [+ or -] 0.481    4.210 [+ or -] 0.389

Group                  Left Hand
                       Pretest                Posttest
RHW (J)
  Low-level training
  Regular training     1,006 [+ or -] 108      1,295 [+ or -]
                                                 134(a,c)
  Control              1,028 [+ or -] 113      1,438 [+ or -]
                                                 101(a,c)
                       1,096 [+ or -] 91       1,114 [+ or -] 87
MVIC (N)
  Low-level training
  Regular training     417.2 [+ or -] 33.42   422.36 [+ or -] 39.97
  Control              385.6 [+ or -] 22.0     401.0 [+ or -] 10.54
                       392.7 [+ or -] 19.2     396.4 [+ or -] 17.91
IHE (rain)
  Low-level training
  Regular training     3.463 [+ or -] 0.192    3.948 [+ or -] 0.392
  Control              4.115 [+ or -] 0.485    4.182 [+ or -] 0.497
                       3.890 [+ or -] 0.392    3.960 [+ or -] 0.410

(a) Posttest condition significantly greater than pretest condition ([F.sub.12,15], P [is less than or equal to] .05).

(b) Posttest condition significantly greater than control and low-level training conditions ([F.sub.2,22], P [is less than or equal to] .05).

(c) Posttest condition significantly greater than control condition ([F.sub.1,15], P [is less than or equal to] .05).

The RHW training effect for either the regular training group (30% of MVIC) or low-level training group (0.005% of MVIC) group was not associated with a corresponding change in MVIC (P = .41) or IHE (P = .37) of the same hand, which suggested a highly specific effect. Figure 1 contrasts the 1,232% increase in RHW (panel A) for the right hand with the 3.8% increase in the same hand's IHE (panel B) and the 8% decrease in the same hand's MVIC (panel C). The lack of a change in MVIC and IHE despite a change in RHW indicates that the training was highly specific.

[Figure 1 ILLUSTRATION OMITTED]

The increase in the ability for the right hand to do work (30% training group) was associated with a corresponding change in the left hand's ability to do work (the average number of repetitions increased from 76.2 to 92.4). The left-hand RHW increased by 43.1% (P [is less than or equal to] .05) following 6 weeks of training the right side (Fig. 1). The low-level training group's right-hand RHW increased 35.6% (repetitions increased from 75.6 to 93.4), an amount similar to the regular training group's untrained hand (left), but greater than the control group's right or left hand (6.44% and 4.42%, respectively). Thus, the magnitude of the cross-education did not appear to depend on the changes associated with training at a 30% load. Instead, training with a low load ([is less than] 0.005% of MVIC) caused increases similar to that observed in the hand opposite to the side that rhymically trained at a 30% load.

The control group showed 6.44% and 4.42% increases in RHW of the right hand and left hand, respectively (average repetitions changed from 77 to 79.8 and 79.6 to 80.4, respectively). The 43.1%, 35.6%, and 32.2% increases in the left hand of the regular training group, the right hand of the low-level training group, and the left hand of the low-level training group, respectively, were greater than the 6.44% and 4.42% increases found for the control group.

Because endurance training was highly specific and led to no change in MVIC or IHE, there was also no cross-education effect for these variables (P = .44) for any of the groups. Thus, endurance training was highly specific to the mode of training.

The average RHW produced each week for the regular training group over the 6-week training protocol is shown in Figure 2. The relationship between right-hand RHW and training duration is best described by the linear regression equation: RHW (J) = 2,164.9 (weeks of training) -- 462.1. Ninety-four percent of the variability in RHW can be explained by the time of training ([R.sup.2] = .94). The "apparent plateau" in RHW for weeks 5 and 6 reflect the 2 subjects who trained for the 2-hour limit as previously described. The increase in the posttest measurements reflects that these 2 subjects were not restricted by the 2-hour time for the final assessment of RHW. Consequently, the mean work during weeks 5 and 6 was excluded when determining the best fit line (diagonal line in Fig. 2).

[Figure 2 ILLUSTRATION OMITTED]

Discussion

Our major findings were that highly motivated subjects can show large increases in RHW as the result of 6 weeks of training, but these changes are specific to the mode of training. That is, rhythmic handgrip training caused no change in maximal isometric handgrip or sustained isometric endurance with a 30% load. Moreover, the RHW of the hand opposite to the side that rhythmically trained at a 30% load did not demonstrate more endurance when compared with the low-level training group. Given that the low-level training group ([is less than] 0.005% of MVIC) did not improve to the same magnitude as the right hand of the regular training group suggests that the low-level training group did not undergo the same adaptations. Large increases in the ability to perform handgrip work has previously been attributed to vascular changes.[12,16] The low-level training group (bilaterally) and the left hand of the regular training group, however, showed a greater improvement in RHW than a control group of subjects who did not train (6.4%). These results suggest that the adaptations associated with aggressive rhythmic handgrip training using submaximal loads (30% of MVIC) do not enhance the opposite limb's ability to perform work beyond what is gained from using the instrumentation at a near-zero load (0.005% of MVIC).

Yasuda and Miyamura[12] found, after doing a similar handgrip training study and measuring blood flow, that peak blood flow in the untrained forearm during a standard fatigue protocol was enhanced following 6 weeks of rhythmic handgrip training and suggested that enhanced blood flow may be the mechanism for cross-education. Two factors should be considered in light of their findings. First, despite finding an increase in peak blood flow in the untrained limb, this untrained limb may also have benefited from what the subjects learned from training the right hand. Second, Yasuda and Miyamura did not examine whether the magnitude of the change in endurance of the untrained fore-arm could also be achieved without inducing an increase in endurance of the trained limb.[12] We addressed this issue, in part, by having a low-level training group that received the opportunity to perform the rhythmic gripping task without using a load that would be expected to stress the muscular system and induce vascular adaptations.[12] Thus, our findings suggest that cross-education during handgrip endurance training may also involve adaptations that are not entirely dependent on the exercise intensity. If our assumption that the low-level training group did not induce vascular adaptations is correct, then blood flow could not have contributed to the cross-education observed in this group. This contention may be in agreement with Sinoway and colleagues,[16] who found that handgrip training was associated with only a localized increase in blood flow of the trained limb.

Effects on Endurance

The 1,232% improvement associated with repetitive handgrip training in our study was extremely high when compared with previous studies of this type. Yasuda and Miyamura,[12] for example, reported a 147% increase in handgrip work following 6 weeks of training using a load equal to 30% of MVIC. Closer evaluation of the endurance training magnitudes for individuals in our study reveals that there was a wide range of improvements, as indicated by the large standard error for the posttraining right-hand RHW of the regular training group ([+ or -] 5,504 J) (Tab. 2). Although all subjects showed at least 100% improvement in RHW of the side trained at a 30% load, 4 subjects showed an increase in RHW of over 1,000% and, therefore, were instrumental in causing the group average to be so high. The discrepancy between the magnitude of the change in RHW in our study and that of Yasuda and Miyamura[12] may be attributed to 2 factors.

First, submaximal repetitive handgripping to fatigue is uncomfortable and requires considerable motivation or "central drive" to perform optimally.[18] Some of our subjects were extremely motivated, and we verbally encouraged our subjects to perform to their maximal effort during each training session. If central fatigue was minimized, motivation to perform maximally would be maximized so that the peripheral components associated with rhythmic muscle training may have been stressed more than in other studies and subsequently may have induced a greater adaptation over the 6-week training period. Peripheral locations involved with fatigue that may have undergone an adaptation include neuromuscular transmission,[19,20] excitation-contraction coupling,[19,21,22] contractile machinery,[19] oxidative metabolism,[19] and blood flow.[19]

Second, our subjects performed 2 bouts of training for the right handgrip during each training session, whereas the subjects in the study by Yasuda and Miyamura[12] performed only one bout of training. The second trial may have been important in inducing the large endurance training effects that we found in the regular training group.

Specificity of Endurance Training

The inability to increase the fight-hand MVIC despite a 1,232% improvement in right-hand RHW is consistent with previous reports.[6,12,23-25] In particular, Yasuda and Miyamura[12] found no change in MVIC despite improvement in RHW, which is consistent with our findings.

Four subjects showed an average increase of 2,676% (SE = 1,507%) in RHW from the 6-week training program, but they showed a 15% decrease (SE = 5%) in MVIC. Conversely, the 4 subjects who showed a less dramatic average improvement in RHW (288%, SE = 181%) showed a 13% (SE = 9.9%) increase in MVIC. These data lend support to the notion that high levels of endurance training may inversely affect MVIC.

Rhythmic handgrip training also did not carry over to IHE. This finding is not surprising, however, because increased blood flow (as well as other adaptations) after 6 weeks of rhythmic handgrip training is thought to be largely responsible for the increase in endurance.[12] Consequently, a continuous static contraction may, in itself, attenuate blood flow and limit the use of newly adapted oxidative machinery induced by the rhythmic handgrip training. Moreover, the mechanisms for fatigue during rhythmic handgrip exercise versus sustained isometric handgrip contractions appear to be different. Fitts[19] suggested that low-intensity, high-duration exercise such as rhythmic handgripping is accompanied by decreases in glycogen and glucose, whereas shorter-duration exercise, such as exercise involving static contractions, is attributed to excitation-contraction coupling failure. Bystrom and Kilbom[23] reported preferential excitation-contraction coupling failure following isometric sustained handgripping (25% of maximum) when compared with an intermittent handgripping group (25% of maximum). Collectively, these reports and the findings of our study support the notion that enhanced endurance is specific to the mode with which training occurs.

When the same training protocol used in our study was carried out using a 50% of MVIC load rather than a 30% load, the MVIC was improved.[12] These findings suggest that training specificity may be partly dependent on the magnitude of the load or the effort exerted during a single contraction. In this context, most researchers reporting cross-education have used protocols involving near-maximal effort.[1,2,4,5,7,10,13,16,26,27]

Cross-Education With Endurance Training

The training of the left hand in the regular training group and both hands of the low-level training group (43%, 36%, and 32%, respectively) were small when compared with the large increase in RHW (1,232%). The low-level training appears to have induced a learning adaptation that was similar in magnitude to the cross-education found in the training group in the study by Yasuda and Miyamura.[12] These findings support the notion that an increased ability to do work in the contralateral limb after rhythmic handgrip training (30% of MVIC) of the ipsilateral limb[12] and the improved ability to do work with low-level training reflect a learning phenomenon that is greater than that observed from a control group that has not received any training. Defining the mechanism of learning as used in this context is difficult because information is lacking. Whether the low-level training group felt more comfortable in the laboratory, believed they were expected to perform better in the posttest, or facilitated synergists and inhibited antagonists are all possible explanations that could not be separated out in this study.

Of the 4 subjects in the regular training group who showed the greatest increases in work (2,677%, SE = 181%), the most dramatic increase occurred in a subject who showed a 46% increase in left-hand handgrip work. Conversely, the 4 subjects in the regular training group who showed the least improvement in right-hand handgrip work (288%, SE = 181%) also had an increase of 35% in left-hand handgrip work. The large magnitude of the differences in the trained limb work between these partitioned groups was not associated with a large difference in the magnitude of the left untrained limb work. If vascular adaptations contributed to the improved ability to perform work in the fight hand and if vascular adaptations contribute to cross-education, then we would have expected a much greater increase in work from the untrained limb of the left hand of the regular training group (30% of MVIC) when compared with either hand of the low-level training group (near-zero load). Accordingly, the mechanism for cross-education during endurance training may primarily involve other adaptations, possibly neural, that can be enhanced through practicing a task at a very low workload. These neural adaptations include any learning that may have occurred at a central or peripheral level of the nervous system.

Conclusion

We found that rhythmic handgrip endurance training (30% of MVIC) caused an increase in RHW, no change in IHE, and no change in MVIC. Based on our work and the work of Yasuda and Miyamura,[12] we suggest that cross-education for improving the ability to do work using a 30% training load involves a mechanism similar to that evoked through practice of a task at a very low workload. Further studies are needed to more clearly define the specific mechanisms leading to cross-education and training specificity during endurance training. As we advance our understanding of cross-education and training specificity, we may discover that these phenomena are important considerations when prescribing exercise for various patient populations, especially patients who are limited by central nervous system dysfunction.

(*) Redington Inc, 222 Lancourt St, Windsor, CT 06095

([dagger]) Genisco Manufacturing, 650 Easy St, Simi Valley, CA 94086.

References

[1] Morrissey MC, Harman EA, Johnson MJ. Resistance training modes: specificity and effectiveness. Med Sci Sports Exerc. 1995;27:648-660.

[2] Housh DJ, Housh TJ. The effects of unilateral velocity-specific concentric strength training. J Orthop Sports Phys Ther. 1993;17:252-256.

[3] Enoka RM. Muscle strength and its development: new perspectives. Sports Med. 1988;6:146-168.

[4] Graves JE, Pollock ML, Forster D, et al. Effect of training frequency and specificity on isometric lumbar extension strength. Spine. 1990;15: 504-509.

[5] Smidt GL, Blanpied PR, White RW. Exploration of mechanical and electromyographic responses of trunk muscles to high-intensity resistive exercise. Spine. 1989;14:815-830.

[6] Fernhall B, Kohrt W. The effect of training specificity on maximal and submaximal physiological responses to treadmill and cycle ergometry. J Sports Med Phys Fitness. 1990;30:268-275.

[7] Behm DG, Sale DG. Velocity specificity of resistance training. Sports Med. 1993;15:374-388.

[8] Kitai TA, Sale DG. Specificity of joint angle in isometric training. Eur J Appl Physiol. 1989;58:744-748.

[9] Hill DW, Cureton KJ, Collins MA. Circadian specificity in exercise training. Ergonomics. 1989;32:79-92.

[10] Moritani T, deVries HA. Neural factors versus hypertrophy in the time course of muscle strength gain. Am J Phys Med. 1979;58:115-130.

[11] Rutherford OM, Greig CA, Sargeant AJ,Jones DA. Strength training and power output: transference effects in the human quadriceps muscle. J Sports Sci. 1986;4:101-107.

[12] Yasuda Y, Miyamura M. Cross-transfer effects of muscular training on blood flow in the ipsilateral and contralateral forearms. Eur J Appl Physiol. 1983;51:321-329.

[13] Sale DG. Neural adaptation to resistance training. Med Sci Sports Exerc. 1988;20:S135-S145.

[14] Hellebrandt FA, Parrish AM, Houtz SJ. Cross-education: the influence of unilateral exercise on the contralateral limb. Arch Phys Med. 1947;28:76-85.

[15] Enoka RM. Neurological Basis of Kinesiology. 2nd ed. Champaign, Ill: Human Kinetics Inc; 1994.

[16] Sinoway LI, Shenberger J, Wilson J, et al. A 30-day forearm work protocol increases maximal forearm blood flow. J Appl Physiol. 1987; 62:1063-1067.

[17] Somers VK, Leo KC, Shields RK, et al. Forearm endurance training attenuates sympathetic nerve response to isometric handgrip in normal humans. J Appl Physiol. 1992;72:1039-1043.

[18] Gibson H, Edwards RHT. Muscular exercise and fatigue. Sports Med. 1985;2:120-132.

[19] Fitts RH. Cellular mechanisms of muscle fatigue. Physiol Rev. 1994;74:49-94.

[20] Shields RK. Fatigability, relaxation properties, and electromyographic properties of the human paralyzed soleus muscle. J Neurophysiol. 1995;73:2195-2206.

[21] Shields RK, Law LF, Reiling B, et al. Effects of electrically induced fatigue on the twitch and tetanus of paralyzed soleus muscle in humans. J Appl Physiol. 1997;82:1499-1507.

[22] Shields RK, Chang YJ. The effects of fatigue on the torque frequency curve of the human paralyzed soleus muscle. J Electromyogr Kinesiology. 1997;7(1):3-13.

[23] Bystrom S, Kilbom A. Electrical stimulation of human forearm extensor muscles as an indicator of handgrip fatigue and recovery. Eur J Appl Physiol. 1991;62:363-368.

[24] Kurosawa Y, Iwane H, Hamaoka T, et al. The effects of endurance training on muscle metabolism in young women by magnetic resonance spectroscopy. Med Sci Sports Exerc. 1995;27(suppl):S44.

[25] Young K, McDonagh MJN, Davies CTM. The effects of two forms of isometric training on the mechanical properties of the triceps surae in man. Pflugers Arch. 1985;405:384-388.

[26] Mayhew TP, Rothstein JM, Finucane SD, Lamb RL. Muscular adaptation to concentric and eccentric exercise at equal power levels. Med Sci Sports Exerc. 1995;27:868-873.

[27] Rutherford OM, Jones DA. The role of learning and coordination in strength training. Eur J Appl Physiol. 1986;55:100-105.

RK Shields, PhD, PT, is Associate Professor, Graduate Physical Therapy Program, College of Medicine, The University of Iowa, 2600 Steindler Bldg, Iowa City, IA 52242-1008 (USA) ([e-mail]). Address all correspondence to Dr Shields.

KC Leo, PT, is Director, Physical Therapy Department, The University of Iowa Hospitals and Clinics, Iowa City, Iowa.

AJ Messaros, PT, was a doctoral student, Graduate Physical Therapy Program, The University of Iowa, at the time of this study.

VK Somers, MD, is Associate Professor, Internal Medicine, The University of Iowa, and recipient of the NIH Sleep Academic Award.

Concept and research design, project management, and fund procurement provided by Shields, Leo, and Somers; writing, by Shields, Leo, Messaros, and Somers; data collection, by Shields and Leo; data analysis, by Shields, Leo, and Messaros; manuscript review prior to submission, by Shields, Leo, Messaros, and Somers. Carol Leigh. provided manuscript preparation.

This study was approved by the Institutional Human Subjects Review Committee of The University of Iowa.

This article was submitted December 30, 1997, and was accepted January 12, 1999.