Muscular exercise causes hyperemia-induced swelling of muscle. This swelling subsides quickly following the exercise. Muscle injury, induced by eccentric exercise, results in swelling which has a delayed onset and a duration of several days. Injury also causes an increase in muscle stiffness, i.e., an increased resistance of the resting muscle to passive lengthening by rotation of the respective limb (13). The degree to which these two variables are causally related following muscle injury is unclear.
Muscle swelling has generally been estimated from the change in limb circumference (3,4,13,18). This measure does not discriminate between muscle swelling and swelling in other compartments. Imaging of the muscle compartment itself either by MRI(18) or by ultrasound, as reported here, provides a more direct way of assessing muscle swelling.
Stiffness has been measured in three different ways. In the human elbow flexors it has been measured as 1) the change in relaxed elbow angle in the standing subject (8-10), 2) the external force required to straighten the elbow (15), and 3) the change in slope of the elbow torque-angle curve of the arm resting in the horizontal plane (13). The first two methods assess the behavior of the arm near full elbow extension, while the latter method, as it has been employed, assesses behavior over the midrange of elbow motion.
The onset of postexercise muscle stiffness as measured by either of the first two methods is gradual, peaking at 3 d postexercise, and exhibiting a time course similar to that of postexercise swelling. Thus, increased stiffness measured in this way may be, at least partially, a result of the initial phase of the swelling. The fact that swelling peaks a day or so later than the effect on the relaxed arm angle may be related to fluid moving from the muscle compartment into the subcutaneous space, where it may have little effect on muscle stiffness.
Clinical observation also suggests that postinjury swelling causes increased stiffness and decreased range of motion. Studies on post-traumatic edema (2) and external pneumatic compression(1) have shown a relationship between swelling and decreased range of motion. Muscle size, unrelated to swelling, is also a determinant of stiffness (6,24).
Stiffness of the elbow flexors following eccentric exercise has a very different time course from swelling if it is measured as the slope of the passive torque-angle curve (13). While arm circumference rises gradually, stiffness rises immediately after the exercise and remains elevated for 4-5 d. Thus, the immediate increase in stiffness measured in this way does not seem attributable to swelling.
The torque-angle curve of passive elbow extension has two distinct portions: an initial portion from 90° to 10° from full extension(Phase 1), and a portion covering the last 10° of elbow extension (Phase 2) (6). The stiffness of Phase 2 is greater than that of Phase 1. Based on Purslow's model of the strain on the perimysial connective tissue, a change in muscle volume would be predicted to shift the Phase 2 portion of the curve without necessarily affecting Phase 1 stiffness(20).
To fully assess the relationship between muscle volume and stiffness, this study was undertaken to look more closely at both stiffness and volume. Both phases of the stiffness curve are examined, and by means of three-dimensional reconstruction of arm compartments from ultrasound images, muscle volume itself, rather than just arm circumference, is examined.
Eleven college age (20.6 ± 1.4 yr) females volunteered for the study. Criteria for the inclusion of subjects in the study were: no weight lifting with the elbow flexors for the 6 months before the study, no history of upper extremity trauma in the past 5 years, not currently taking any antiinflammatory medications, and agreement not to begin regular exercise with the elbow flexors or begin taking antiinflammatory medication during the study. All subjects were oriented to the experiment and signed a consent form approved by the Ohio University Institutional Review Board.
Before eccentric exercise, isometric maximal voluntary contraction (MVC) of the elbow flexors was determined with the subject seated on a bench, which supported the axilla and upper arm at 30° of shoulder flexion. The elbow angle was 90°. MVC was taken as the maximum of three repetitions of isometric elbow flexion effort (resting 2 min between each contraction) on three different days during the week before the bout of eccentric exercise. The subjects performed three sets of eccentric exercise with the elbow flexors of their nondominant arm. Loads used in the three sets were 90, 80, and 70% of the isometric MVC. By means of a padded mechanical arm whose axis of rotation matched that of the elbow, the load was applied perpendicular to the distal forearm throughout the range of motion. The subjects were instructed to slowly lower the weight over a count of 8 s. As the subjects began to fatigue, they were no longer able to slowly lower the weight near full elbow extension. The subjects were encouraged to try lowering the weight 1-2 more times with an emphasis on the need to control the descent of the weight. When they were no longer able to control the descent of the weight, the set was terminated, and the subject was allowed to rest for 2 min to minimize the effects of fatigue before beginning the next set. Typically, the subjects performed only 4-6 repetitions in each set before they could no longer control the descent of the weight.
Measurements of muscle soreness, isometric strength, stiffness, and swelling were made three times before exercise, immediately after exercise(“post”), daily for 5 d, and every other day during the second week after exercise. Soreness was measured using a five-point scale: 0) no pain, 1) pain on palpation only, 2) mild pain with full flexion or extension of the elbow, 3) significant pain with full flexion or extension of the elbow, and 4) constant pain. Isometric strength was measured as described above for determining isometric MVC.
For measurements of stiffness, an apparatus that was described previously was used with a few modifications (6,7,13). With the subject sitting, the shoulder was stabilized in a position of 90° abduction and 15° horizontal extension, and the upper arm was stabilized on a stationary platform. The forearm was secured to a second platform which articulated with the first at a point corresponding to the axis of the elbow joint. This platform was moved by a stepper motor in 4° steps from a resting position (90°) to full elbow extension. Extension was stopped when the humerus began to visibly rotate externally in response to full extension of the elbow. The system was programmed to record angle and torque 12 s after each step to allow for 80-90% of stress relaxation. The slopes of the initial linear portion (Phase 1) and the second steeper portion (Phase 2) of the torque-angle curve (Fig. 1) and the break point (the angle where Phase 1 ends and Phase 2 begins) were calculated(16). The mean slope (in Nm·deg-1) of three pre-exercise stiffness measurements was used as a reference for all postexercise stiffness measurements. By putting the shoulder in horizontal extension, the biceps muscle (which crosses the shoulder and the elbow joints) will be strained to a greater extent during elbow extension, and tension will be placed primarily on the elbow flexor muscles and tendons, not the joint structures. Surface electromyographic (EMG) recordings of the biceps and triceps were inspected visually at the time of the stiffness measurement to ensure that the subjects were relaxed. Stiffness and EMG data were recorded with a microcomputer using the RC Computerscope data acquisition package (RC Electronics Inc., Santa Barbara, CA). EMG data were later processed using the rectified, averaged EMG during the stiffness measurement. A second measure of stiffness used was relaxed arm angle. With the subject standing we measured the angle of the elbow with a standard plastic goniometer using the ulnar styloid, lateral epicondyle, and the anterior lip of the acromion as landmarks.
Change in arm circumference and muscle compartmental volume were used to measure the amount of swelling in the upper arm. Circumferential measurements were taken with a Gulick tape measure at the mid belly of biceps and at a point 2.5 cm proximal to the lateral epicondyle. The points were marked on the subject's arm to ensure consistent placement of the tape measure. Muscle compartmental volume was measured with a custom ultrasound imaging unit designed and built by the Ohio University Department of Electrical and Computer Engineering. The unit consisted of a water tank, a B-mode diagnostic ultrasound unit, a device to move the sound head, and a computer. The water was kept at 30-32°C, and the subject's arm was immersed in a vertical position, with the elbow flexed at 90° and the forearm secured to a horizontal platform. The sound head of the ultrasound unit was attached to a computer controlled device that moved circumferentially around the arm recording several images directly to the computer. The images were then compounded by the computer to create a cross-sectional image of the arm. After completing the circumferential movement at one level, the device moved longitudinally 1 cm and another cross-sectional image was recorded. This was repeated to include the whole upper arm from the axilla to about 1 cm proximal to the cubital fossa. Figure 2 shows representative images from one subject who had a 30% increase in volume of the elbow flexor compartment following the exercise. The images were processed manually using a computer mouse to outline the circumference of the arm, the elbow flexor compartment (biceps and brachialis), the triceps compartment, and the humerus. The area of each compartment was then calculated by the computer for each cross-sectional image. Volume of each compartment was calculated as the sum of the cross-sectional area (CSA) across slices since the distance between each slice was 1 cm. An estimate of the subcutaneous volume was derived by subtracting the sum of the volume of the flexor and extensor compartments and the humerus from the total arm volume. The accuracy of the volumes of test objects estimated from ultrasound images compared with that of their actual volumes was 2.2% ± 1.1% for simulated muscle and 6.9% ± 2.3% for simulated bone. The coefficient of variation was 1% for test objects and 1.75% for the elbow flexor compartment. Intra-rater (ICC (3,1)) and interrater (ICC(2,1)) reliability were both found to be 0.99, based on the intraclass correlation coefficient (ICC) (6). The correlation (r) between the two pre-exercise scans was 0.94, and a paired t-test showed no difference between the scans (P = 0.81).
The time course of each variable except soreness was analyzed with repeated measures ANOVA to determine main effects across time after the exercise. Tukey's post-hoc test for multiple comparisons was used to determine significant differences from pre-exercise values when significant main effects were found. All data are presented as the mean percentage change (± SE) from pre-exercise values.
Soreness followed the familiar pattern of delayed onset muscle soreness, peaking on days 2 and 3 after exercise. The average peak soreness rating was 2.9 on day 2, which means that in general the subjects had significant pain only with movement of the elbow and not while at rest. Isometric MVC decreased immediately after exercise and reached a nadir on postexercise day 1 (47.0± 4.2%, P < 0.01, Table 1). Strength gradually returned but remained significantly decreased 11 d after exercise(P < 0.01). A previous study demonstrated that full strength recovery occurred but required as long as 12 wk (13).
Minimal swelling of the elbow flexor compartment was observed immediately after exercise; swelling then increased over postexercise days 1 and 2, was maintained over days 3-5, and gradually decreased over days 7-11(Fig. 3). The volume of the flexor muscles increased by as much as 26.1 ± 4.3% after exercise with days 2-7 exhibiting significant increases (P < 0.05) over pre-exercise volumes. Swelling of the whole arm showed a similar time course as the flexor muscle swelling, but the percentage change was less. The flexor volume as a fraction of the whole arm volume was 15.6% before exercise, and 19.1% on day 3 after exercise. The subcutaneous tissue showed a trend (mean increases were not significant) toward increased volume, but the apparent change in volume tended to be slightly delayed compared with that of the flexor muscle and arm volumes(Fig. 3). As expected, the triceps volume and the volume of the humerus do not change after exercise (P = 0.16 andP = 0.37 respectively). The circumference, as measured with a tape measure, at both the mid and distal arm increased, reaching a peak change of about 6% on day 4 (Table 1).
Cross sections at two points on the arm were analyzed to determined whether the patterns seen in the volume measurements were specific to particular locations. The most distal cross section was chosen because it is here that the most obvious swelling occurs based on visual examination of the subjects' arms. At this level the flexor compartment CSA, like the volume, significantly increased beginning on postexercise day 1, whereas swelling in the subcutaneous tissue was significant but delayed (Fig. 4). At the cross section 2 cm proximal to the origin of the brachialis muscle, the time course of biceps swelling was similar to the time course of the swelling in the flexor compartment, which also includes the brachialis(Table 1). The biceps increased significantly on days 2-7 with maximal swelling on postexercise day 3 (24.2 ± 5.2%). None of the other compartments at this cross section had significant increases in area.
The slope of Phase 1 of the torque-angle curve, indicative of stiffness, increased by 59.9 ± 14.1% (P < 0.05) immediately after exercise and remained elevated for 5 d (Fig. 5). The slope of Phase 2 was unchanged following the exercise. The break point, or the angle at which Phase 1 ends and Phase 2 begins, shifted to smaller angles(increased elbow flexion) with angles at days 3-7 being significantly less(P < 0.05) than the pre-exercise mean (Fig. 6). Like the breakpoint, the relaxed arm angle became more flexed after exercise (Fig. 6).
The single episode of eccentric exercise resulted in the typical symptoms of delayed onset muscle soreness, strength loss, swelling, and increased stiffness described by others (8,11,13), each with a distinct time course. Soreness followed the characteristic pattern of peaking 2 to 3 d after the exercise and returning to pre-exercise levels by day 7 after exercise. Strength fell immediately after the exercise, continued falling the next day, and then returned gradually, but in some cases not for weeks.
Muscle swelling, as measured with ultrasound imaging, was substantial (26%) in the elbow flexor compartment. Extracellular edema is likely related to tissue damage and the inflammatory response. Mast cell degranulation 48 h postexercise with the release of histamine may cause vasodilation and increased vascular permeability in the area of the injury(22). Prostaglandin E2 synthesis also increases within the first 24 h after injury and may further increase vascular permeability (21). The increased vascular permeability may allow plasma proteins to leak into the interstitial spaces, further increasing interstitial colloid osmotic pressure (22). Interstitial osmotic pressure may be influenced by creatine kinase and other proteins released from damaged cells. The time course of creatine kinase appearance in the blood after eccentric exercise-induced injury reflects a time course similar to swelling, having a gradual onset and a peak about 4-5 d after exercise (9). Fritz and Stauber(12) identified several changes in the local proteoglycan composition of muscle injured by lengthening contractions that could lead to increased binding of water by proteoglycans and swelling in the area of the injured fibers. All these mechanisms have delayed time courses suggesting that they could be responsible for the delayed rise in flexor compartment volume.
Postexercise swelling in humans has two observable phases. Palpation indicated that the exercised muscle was initially firm and enlarged. After about 3 d postexercise the area became soft to palpation, giving the impression of fluid within the subcutaneous tissue. The data from the cross-sectional ultrasound images taken just proximal to the elbow joint reflected this observation. The flexor compartment (biceps and brachialis) swelled quickly, CSA increasing 15% on day 1, and reaching maximum swelling(27%) on day 3. The CSA of the subcutaneous tissue did not begin to increase until day 3 and peaked on days 4 and 5 (8%). This delay in the subcutaneous swelling may represent the movement of fluid from the muscle compartment to the subcutaneous tissue. The swelling was undoubtedly accompanied by increased compartmental pressure as reported by Friden et al.(11). The swelling within the flexor compartment is likely a combination of intracellular and extracellular edema. This was shown in a study using rats that ran on a treadmill with a positive incline; fiber and nonfiber area increased within the first 24 h after exercise(19). Extracellular swelling was found in rabbit triceps surea after an eccentric exercise protocol (4). In humans Friden et al. (11) found that eccentric exercise of the tibialis anterior muscle in humans resulted in increased muscle fiber area 48 h after exercise. Whereas extracellular edema arising from tissue injury results primarily from protein redistribution across capillary membranes, intracellular edema results from electrolyte redistribution across cell membranes. Cell swelling occurs when ATP-dependent Na+ extrusion cannot keep up with passive Na+ influx, either because of excessive leakiness of the cell membrane or because of compromised Na+ pump function, both of which may occur as a result of injury.
Because we cannot measure muscle strain directly in human subjects, we measured the elbow angle and used the slope of the passive torque-angle curve to provide our estimate of stiffness. As shown in Figure 1 the curve has two distinct components, suggesting that two mechanical processes contribute to stiffness. Phase 1 may be determined by factors intrinsic to the muscle, such as residual cross-bridge attachments(17) or the elastic connecting protein titin and the cytoskeleton (23). In addition, Phase 1 stiffness has been shown to be correlated to the volume and cross-sectional area of the muscle (6,24). In intact muscle the second steeper portion of the curve, Phase 2, may be determined by extracellular connective tissue (20) and joint structures such as the capsule and ligaments (14).
After eccentric exercise the slope of Phase 1 increased, the break point shifted to more acute elbow angles, but the slope of Phase 2 did not change. Resting elbow angle also provided a measure of the stiffness. The resting elbow angle became more flexed after exercise, also representing an increase in muscle stiffness. Are either of these changes in the stiffness a reflection of the change in muscle volume?
The time courses of the shift in the break point and resting arm angle were very similar to the change in volume of the flexor compartment(Figs. 3 and 6). The rise in volume of the muscle compartment may increase tension on the connective tissue of the endomysium and perimysium surrounding the muscle, resulting in the shift of the angle at which their contribution to the stiffness curve begins. In other words, the elbow becomes more flexed and the break point shifts to the left. In Purslow's model (20) either extreme lengthening or shortening of the muscle increases the tension on the collagen fibers, causing them to stretch by uncrimping. In swollen muscle, as in the present experiments, collagen fibers will come under tension at shorter muscle lengths as the muscle is lengthened. In our experiments this may be reflected by a shift in the resting arm angle and the break point to shorter muscle lengths.
Although the time courses of the breakpoint and relaxed arm angle matched the changes in flexor compartment volume, neither phase of the stiffness curve did. The change in slope of Phase 1 preceded the change in volume of the muscle compartment. The slope increased immediately after exercise and remained elevated for 5 d (Fig. 7). The decline in Phase 1 stiffness after day 5, however, did parallel the decline in compartmental volume. This suggests that the swelling cannot account for the initial increase in stiffness, but the continued stiffness and the return to baseline may be related to the swelling. The absence of change from pre-exercise in the slope of the second phase of the curve suggests that the structures responsible for this portion of the curve, such as the joint capsule and ligaments, are unaffected by the swelling within the muscle compartment.
The question of what accounts for the immediate rise in Phase 1 stiffness remains. This portion of the curve comprises most of the range of elbow extension from the resting angle, 90-100°, to the break point, which is 8-10° from full extension. An increase in resting EMG could in principle account for the increased tension, but EMG levels before and after exercise were not different (biceps P = 0.47; triceps P = 0.27). It has been hypothesized that low level calcium activation could be responsible for the increased slope (13). This effect would have to be length dependent, greater at longer lengths and less at shorter lengths, since it is the slope that increases. If it were simply the addition of calcium in the muscle cell causing increased tension regardless of fiber length, the entire curve would shift to higher levels of tension, but the slope would be unchanged. The length dependency of the tension response has been shown in vascular smooth muscle cells where osmotically-induced stretch of the membrane caused increased contractile activity(5).
Despite the time courses of swelling and of Phase 1 stiffness being different, there is some evidence that the two variables are related. When external pneumatic compression was applied daily to the arm after exercise-induced muscle injury, the circumference of the arm and the slope of Phase 1 decreased on days 2 and 3 after exercise (7). This suggests that the swelling may be responsible for a portion of the increased Phase 1 stiffness during the time that the muscle compartmental volume was increasing.
This study has detailed the time course of elbow flexor muscle compartment swelling resulting from muscle injury after eccentric exercise. The delayed onset of swelling cannot account for the immediate rise in muscle stiffness. This increased stiffness may be related to the change in calcium homeostasis as a result of the muscle injury. Although the onset of stiffness does not coincide with swelling, the dissipation of the stiffness does coincide with the return of the flexor muscles to normal volume. This suggests that there-turn of stiffness to normal values may be a function of muscle edema.
The authors thank Aaron Brumit, P.T., Tom Campbell, P.T., Scott Johnson, P.T., Sam Simons, P.T., Jason Strong, P.T., and Trent West, P.T. for many hours of assistance in data collection.
This work was supported in part by the Foundation for Physical Therapy, Inc., and the Bureau of Research, American Osteopathic Association.
Address for correspondence: Gary S. Chleboun, Ph.D., P.T., Ohio University School of Physical Therapy, 199 Convocation Center, Athens, Ohio 45701. E-mail: firstname.lastname@example.org.
1. Airaksinen, O. Changes in posttraumatic ankle joint mobility, pain, and edema following intermittent pneumatic compression therapy. Arch. Phys. Med. Rehabil.
70: 341-344, 1989.
2. Airaksinen, O., K. Partanen, P. J. Kolari, and S. Soimakallio. Intermittent pneumatic compression therapy in post-traumatic lower limb edema: computed tomography and clinical measurements. Arch. Phys. Med. Rehabil.
72: 667-670, 1991.
3. Bobbert, M. F., A. P. Hollander, and P. A. Huijing. Factors in delayed onset muscular soreness of man. Med. Sci. Sports Exerc.
4. Brendstrup, P. Late edema after muscular exercise.Arch. Phys. Med. Rehabil.
43: 401-405, 1962.
5. Bulow, A. and B. Johansson. Membrane stretch evoked by cell swelling increases contractile activity in vascular smooth muscle through dihydropyridine-sensitive pathways. Acta Physiol. Scand.
6. Chleboun, G. S., J. N. Howell, R. R. Conaster, and J. J. Giesey, The relationship between elbow flexor volume and angular stiffness at the elbow. Clin. Biomech.
, 12:383-392, 1997.
7. Chleboun, G. S., J. N. Howell, H. Baker, et al. Intermittent pneumatic compression effect on eccentric exercise-induced swelling stiffness and strength loss. Arch. Physiol. Med. Rehabil.
8. Clarkson, P. M. and M. E. Dedrick. Exercise-induced muscle damage, repair, and adaptation in old and young subjects. J. Gerontol.
9. Clarkson, P. M. and I. Tremblay. Exercise-induced muscle damage, repair, and adaptation in humans. J. Appl. Physiol.
10. Ebbeling, C. B. and P. M. Clarkson. Muscle adaptation prior to recovery following eccentric exercise. Eur. J. Appl. Physiol.
11. Friden, J., P. N. Sfakianos, A. R. Hargens, and W. H. Akeson. Residual muscular swelling after repetitive eccentric contractions.J. Orthop. Res.
12. Fritz, V. K. and W. T. Stauber. Characterization of muscles injured by forced lengthening. II. Proteoglycans. Med. Sci. Sport. Exerc.
13. Howell, J. N., G. S. Chleboun, and R. Conatser. Muscle stiffness, strength loss, swelling and soreness following exercise-induced injury in humans. J. Physiol.
14. Johns, R. J. and V. Wright. Relative importance of various tissues in joint stiffness. J. Appl. Physiol.
15. Jones, D. A., D. J. Newham, and P. M. Clarkson. Skeletal muscle stiffness and pain following eccentric exercise of the elbow flexors
16. Jones, R. H. and B. A. Molitoris. A statistical method for determining the breakpoint of two lines. Analytical Biochem.
17. Lakie, M. and L. G. Robson. Thixotropy in frog single muscle fibres. Exp. Physiol.
18. Nosaka, K. and P. M. Clarkson. Changes in indicators of inflammation
after eccentric exercise of the elbow flexors
. Med. Sci. Sports Exerc.
19. P. Binkhorst, F. M., D. W. Slaaf, H. Kuipers, G. J. Tangelker, and R. S. Reneman. Exercise-induced swelling of rat soleus muscle: its relationship with intramuscular pressure. J. Appl. Physiol.
20. Purslow, P. P. Strain-induced reorientation of an intramuscular connective tissue network: implications for passive muscle elasticity. J. Biomech.
21. Smith, L. L. Acute inflammation
: the underlying mechanism in delayed onset muscle soreness. Med. Sci. Sports Exerc.
22. Stauber, W. T., V. K. Fritz, D. W. Vogelbach, and B. Dahlmann. Characterization of muscles injured by forced lengthening. I. Cellular infiltrates. Med. Sci. Sports Exerc.
23. Wang, K., R. McCarter, J. Wright, J. Beverly, and R. Ramirez-Mitchell. Viscoelasticity of the sarcomere matrix of skeletal muscles.Biophys. J.
24. Wiegner, A. W. and R. L. Watts. Elastic properties of muscles measured at the elbow in man: I. normal controls. J. Neurol. Neurosurg. Psychiatry
Keywords:© Williams & Wilkins 1998. All Rights Reserved.
ELBOW FLEXORS; ULTRASOUND IMAGING; INFLAMMATION; MUSLCE EDEMA; ELBOW JOINT; MUSCLE INJURY