Chronic exertional compartment syndrome (CECS) is a common cause of exercise-induced lower-leg pain, which most often affects the anterior and deep posterior compartments (1,13,20).
It is a transient and reversible condition triggered by activity and relieved by rest. CECS patients can experience considerable pain and temporary neurological deficit (10), and in many cases, the pain is severe enough to restrict exercise and potentially affect career prospects, e.g., in service personnel and competitive athletes.
Although not a universal finding, several authors (5,12,13,18,19,20) have demonstrated abnormally elevated intramuscular pressures during exercise in CECS patients, and these also are associated with reduced muscle blood flow (2,19,21,23,25).
Turnipseed et al. (25) investigated the effects of increased compartment pressure on anterior tibial arteriovenous flow patterns and reported normal arterial flow but impaired tibial vein drainage in CECS patients. No venous occlusions or anatomical abnormalities were found, so reduced flow was attributed to the increased venous pressure. The fascia of the affected compartment in the patients was also found to be significantly thicker and stiffer than that taken from unaffected compartments and from control subjects. During exercise, it is normal for muscle volume to increase by up to 20% via increased capillary perfusion, blood volume, and interstitial edema and muscle fiber swelling (10). It is possible that in an abnormally inextensible compartment, the tissue swelling that is a normal consequence of metabolic activity during exercise leads to an increase in pressure sufficient to compromise tissue perfusion (5,8,10,12,13).
Although widely accepted, there is little direct evidence for these suggestions of the cause of CECS, specifically the mechanisms whereby the circulatory impairment occurs, i.e., some direct changes in the venous circulation or an indirect effect through increased tissue pressures. If the former were the cause, then the expectation would be that the syndrome could be simulated in healthy people by external occlusion of venous flow to exercising muscle. Therefore, we have compared the response to exercise in terms of muscle function, thickness, and pain in CECS patients with normal controls. The latter performed two bouts of exercise, one with a free circulation and the other with partial occlusion.
Ten healthy subjects (five males, five females) with no previous history of circulatory disease or musculoskeletal conditions and 20 CECS patients (16 males) participated in the study. Inclusion criteria for the patients was a medical history compatible with CECS and positive intramuscular pressure measurements (mean exercise pressure during symptomatic exercise of >50 mm Hg, >30 mm Hg at 1 min after exercise, and >20 mm Hg 5 min after exercise. Other causes of lower leg pain associated with exercise (e.g., stress fractures and periostitis) were eliminated by x-ray and bone scan.
All subjects gave written informed consent to participate in the study that was conducted in accordance with the Declaration of Helsinki (1989) of the World Medical Association. It had been approved by the local ethics committees.
The subjects were seated on a dynamometer with the backrest adjusted to an angle of 60° and the knee joint at 45° of flexion. The foot was positioned in 10° of plantarflexion (0° corresponded to a perpendicular relationship between the tibia and sole of the foot).
Kincom II and Biodex (system 3) dynamometers were used. The compatibility of the two systems was previously determined by conducting repeat measures on both machines in five control subjects. Measurements of both voluntary and stimulated force were shown to agree (P > 0.05, paired t-test), and no bias was observed (6).
After familiarization with the equipment and procedure, subjects were asked to perform brief (2-s) maximal isometric voluntary contractions (MVC) of the anterior tibial (AT) muscles with visual and verbal feedback.
The AT muscles were electrically stimulated at 50 Hz for 2 s (pulse width 0.1 ms) via two carbon-rubber adhesive pads (4.0 × 4.0 cm) placed as far apart as possible on the skin overlying the muscle. The stimulus intensity was the maximum tolerated, providing that a level plateau was generated (3).
The MVC and tetanic forces were recorded before exercise. During the exercise protocol, tetanic force was measured at 60-s intervals. During recovery, voluntary and stimulated force was recorded 1, 3, 5, 7, and 10 min after the end of exercise.
The subjects were asked to perform isometric MVC for 1.6 s with a duty cycle of 0.5 for 20 min, using an audio signal for timing. The force trace, displayed on a PC monitor, provided visual feedback that was augmented by verbal encouragement. Subjects remained seated in the dynamometer during the 10-min recovery period.
Force signals (sampling frequency 1 kHz) were A to D converted (Cambridge Electronic Design (CED), UK), then displayed and recorded for subsequent analysis using Signal software (CED).
Control subjects performed the protocol on two occasions at least a week apart. In one session, it was performed with a free circulation. In the other, a sphygmomanometer cuff was fitted just below the knee and inflated to 80 mm Hg. This pressure was maintained until the end of exercise and released for the recovery period.
A preliminary study of blood flow in the anterior tibial vein using Doppler ultrasound indicated that a pressure of 80 mm Hg was sufficient to arrest blood flow in the tibial vein in normal subjects (N = 5) at rest. This pressure was below the diastolic pressure at the ankle when the subject was seated in the dynamometer, and therefore arterial inflow to the muscle was maintained. Patients performed the exercise and test protocols with a free circulation on one occasion only.
Subjects were asked to rate verbally any pain or discomfort in the AT muscles on a 0–10 scale (0 = no pain, 10 = worst pain imaginable) before exercise, at 1-min intervals during and then at 1, 3, 5, 7, and 10 min after exercise. They were specifically asked to rate pain in the exercising AT muscle and not that under the sphygmomanometer cuff. The subjects reported that they had no difficulty in distinguishing between the two.
Real-time ultrasound imaging was used to measure muscle thickness before exercise and during recovery, using a portable B-scanner (Aloka SSD-900) with a 7.5-MHz linear probe. The scan site (20% of the distance from the head of the fibula to the lateral tip of the lateral malleolus (14)) was marked on the skin. Coupling gel was spread over the skin and the transducer. The transducer was held at an angle where the interosseous membrane was parallel to the top of the scanner screen.
The images were recorded on video and individual frames saved onto PC using a video capture card and specific software (Win-TV, Hauppauge Computer Works, Inc.). The muscle depth measurements were made using Scion image Software (Scion Corporation). Muscle depth was measured from the interosseous membrane to the outer border of the muscle group at a point 10 mm lateral to the tibia.
Analysis of Data
Preexercise measures of force and muscle thickness for the two control conditions were compared using paired t-tests. The percentage changes from the initial value were determined and then an arc sine transformation was used to fit the linear regression model. The effects of time and occlusion on force within subjects were tested using repeated measures ANOVA. One-sample t-tests were used to investigate the values after 1 min of exercise with the preexercise values and the changes in muscle thickness at the end of exercise. Changes in muscle thickness between the two conditions were compared with a paired t-test.
Pain scores were ranked within each subject across the two conditions. The pain scores obtained for each subject were ranked in ascending order and then the ranked data analyzed with a repeated measure ANOVA to investigate the effects of time and occlusion.
Independent samples t-tests were used to compare the physical and preexercise measures of voluntary and stimulated force and muscle thickness.
The patient force and muscle thickness data also underwent an arc sine transformation. The patient data were compared with the controls with free circulation and the controls with venous occlusion separately, using repeated measures ANOVA. Four of the patients were unable to complete the full 20 min of exercise and stopped after 14–18 min. The data from these subjects were not initially included in the analysis. Subsequently, an analysis was conducted on all 20 patients, but only the first 14 min of exercise was considered. Changes in muscle thickness after exercise for all the patients and controls were compared using an independent samples t-test.
The four patients that failed to complete 20 min of exercise were assigned a pain score of 11 for the remaining exercise period. Assigning the number 11 to the missing data enabled these patients’ results to be included in the analysis and took into account the severity of their pain that prevented them from continuing the exercise. One patient who completed 20 min of exercise subsequently experienced numbness and was unable to rate pain, and so those data have been excluded.
The pain data were analyzed in two ways. Initially the scores were ranked within subjects (as above) and repeated measures ANOVA was performed on the ranked data so that the effects of time and the time × group interactions could be determined.
To determine any difference between the patients and controls in the two conditions, the data were then ranked across groups at each time point from the preexercise measure to the end of exercise, thus removing time from the analysis. A repeated measures ANOVA was then performed.
Transformed data were converted back into the original units for clarity in text and figures. Statistical significance was set at P < 0.05. Data are presented as group mean ± SEM unless stated otherwise.
The physical characteristics of the subjects are shown Table 1. The patients were heavier (P = 0.001) and had a higher body mass index (P = 0.003) than the control group, otherwise there were no significant differences between them. As the control subjects performed the protocol with and without venous occlusion, we refer to three groups (controls with and without occlusion and the CECS patients).
The initial voluntary and stimulated forces were similar in all the patients and controls (Table 1).
The control subjects showed an initial rapid loss of voluntary force (Table 1;Fig. 1a) with a significant force loss after 1 min of exercise in both conditions (P < 0.001). Thereafter, there was a lower rate of fatigue that tended to plateau after 5 min at about 70% of initial force with a free circulation and 60% with an occluded circulation. The latter group showed a greater force loss (P = 0.004).
Recovery was similar in both control groups, with voluntary force returning to approximately 85% of initial values after 1 min.
The patients’ MVC declined to 55% of the initial value over the 20 min of exercise (Fig. 1a). During the first 5 min of exercise, the patients’ MVC were similar to the unoccluded controls and significantly greater than controls with venous occlusion (P < 0.05). The reverse was observed during the last 5 min of exercise, where the patients’ MVC were similar to controls with venous occlusion and significantly less than controls with free circulation (P < 0.05). Recovery was rapid in the patient group, reaching 72% of initial values after 1 min and similar to the recovery pattern observed for both of the control groups.
The initial tetanic forces of all three groups were similar (Table 1;Fig. 1b).
Stimulated force (Fig. 1b) with an intact circulation showed a small but steady decline, reaching approximately 90% after 20 min (P < 0.005). In contrast, with external occlusion, there was a biphasic response, with a rapid force decline in the first 2 min of exercise (P < 0.005) but thereafter a relatively constant force. The force produced under external occlusion was significantly lower at the end of exercise (P = 0.005). In both cases, stimulated force recovered to approximately 95% of initial values after 1 min.
The pattern of fatigue was similar for patients and controls with free circulation. Stimulated force declined to 73% of initial values by the end of exercise in the patients. Their pattern of force loss was intermediate between those of the controls with and without occlusion, and forces were always reduced more than those of the normal subjects with a free circulation. During the first 5 min of exercise, the patients showed less tetanic force fatigue than the occluded controls (P < 0.05), but thereafter there were no differences between the two groups. During the last 10 min of exercise the rate of fatigue in the two groups was very similar.
Tetanic force recovered to approximately 96% after 1 min in the patients, a pattern similar in all the groups. In all three groups the decline in voluntary force was greater than that of tetanic force, indicating the presence of voluntary activation failure. Similar results were also obtained for voluntary and stimulated force when the analysis was conducted on only the first 14 min of exercise and included the four patients who stopped because of pain.
There was a significant increase in muscle thickness after exercise in the control group under both blood flow conditions (P < 0.0005) (Table 1 and Fig. 2). This increased by 15.5 ± 0.1% with external occlusion, more than the increase seen (P = 0.001) when exercising with an intact circulation (4.8 ± 0.1%). During the recovery period, muscle thickness remained significantly greater than the initial measure (P = 0.012–0.0005) and was similar in both control conditions.
Muscle thickness after exercise showed an increase (P < 0.0005) of 6.2 ± 0.2%. This was similar to that found in controls without external occlusion but less (P < 0.05) than after exercise with external occlusion.
Muscle thickness had not returned to initial values after 10-min recovery. There was a nonsignificant trend toward slower recovery in the patients compared with the controls with free circulation.
The pain scores in both conditions increased significantly over time (P < 0.0005), but the two conditions showed a significantly different pattern of increase (P < 0.0005) (Table 1 and Fig. 3). With external occlusion, pain increased rapidly during the first 10 min of exercise and then plateaued at around 13 min. However, with a free circulation, pain increased more gradually and reached a lower plateau later in the exercise period.
Figures 3a and b show the time course of changes in ranked pain. When paired observations were made, the control subjects reported more pain during exercise with venous occlusion (P < 0.0005). At the end of the exercise, the median pain score was 4.8 (range 2–10) in the occluded condition compared with 2.5 (range 0–9.5) with free circulation.
Pain levels declined rapidly during the recovery period (P < 0.0005) and reached preexercise levels after 10 min. There was no significant difference between the two conditions once the cuff had been removed.
There was a significant increase in pain over the exercise period (P < 0.0005). It increased much more rapidly than in the controls with free circulation (P < 0.0005) but similarly to the controls with venous occlusion (Fig. 3b).
When the pain scores for the groups were ranked at each time point across the groups (patients compared separately to each control group), time was not a factor (as in Fig. 3) and only between group differences were considered. These data showed that the patients reported similar pain levels during exercise to controls with occlusion (P = 0.204) but significantly greater pain than the controls with free circulation (P = 0.01). After 20 min of exercise, the median pain score for the patients that completed the exercise (N = 15) was 6 (range 1–10) (Fig. 3c).
Pain declined significantly during the recovery period (P < 0.0005) and returned to preexercise levels. The pattern during recovery was similar in all three groups.
These data largely confirm our hypothesis that partial obstruction of the venous outflow of normal muscles would reproduce the symptoms experienced by CECS patients. Although the patients’ contractile performance in the initial stages of exercise was similar to control subjects with an intact circulation, toward the end, they more closely resembled those with external occlusion.
It appears that there are changes in the working muscles of CECS patients, occurring with a time course of 5–10 min, which lead to an obstruction of the venous outflow and result in pain and fatigue. This would suggest vascular competency at rest and during the early phases of exercise but a progressive failure as exercise continues.
Assuming that the cause of the progressive force failure during exercise in the patients is the build-up of intramuscular pressure, the present results are consistent with the notion that the anterior tibial compartment is less compliant than normal. If the similar rates of decline in force in the latter part of exercise are due to the same thing, i.e., build-up of intramuscular pressure and partial ischemia, then similar pressures would be expected in the two situations. The findings of much greater thickness changes in the occluded normals than patients argue for reduced compartment compliance in the patients. Direct or indirect measures of compartment pressure changes during exercise and recovery should be compared in these three groups to finally resolve this issue.
In the case of the normal subjects with occluded circulation, the electrically stimulated force declined rapidly in the first 2 min only to partially recover before declining again but at a slower rate. This may represent a delayed vascular response with an initial ischemic period at the start of exercise, followed by a period of relative hyperemia and a subsequent steady state during which time oxygen delivery nearly, but not quite, matched the requirements of the tissue.
It is clear that the electrically stimulated forces fell more than the voluntary ones, revealing a failure of voluntary activation. This was of a similar magnitude in all groups, despite the different pain levels of pain reported and emphasizes the need for measures of muscle strength that are not influenced by motivation or central factors.
In CECS patients and control subjects with free circulation there was an increase in muscle thickness of approximately 5%, similar to the changes reported by Gershuni et al. (9) in normal subjects exercising to fatigue. Much greater thickness increases of approximately 15% were found with external occlusion in control subjects. Therefore, this compartment is capable of distension in CECS patients but not necessarily to the extent necessary during exercise. If fascial compliance were lower in the patients, then smaller changes in muscle thickness would be expected after exercise. However, this was not the case, and thickness changes were similar in the patients and controls without external occlusion.
With muscle activity, there is a shift of fluid into the interstitial spaces (22). Normally, the fascia is sufficiently compliant to accommodate this without any clinically significant increase in intramuscular pressure (IMP). If the fascia is abnormally thick or stiff, then the movement of fluid will cause a rise in IMP that will build up and eventually impair the circulation and cause pain and fatigue.
Although the effect on muscle performance was similar in both patients and control subjects with occlusion, the source of the pressure may be different. In occluded control subjects, most of the fluid would presumably accumulate in the blood vessels (22), but with CECS patients, it may be in the interstitial space.
IMP measurements are necessary to fully interpret these data, but if, as generally thought, the patients develop high IMP during exercise, then the differences in muscle volume between patients and control subjects with occlusion probably reflects differences in compliance of the anterior tibial compartment. CECS patients are reported to have a thick and inextensible fascia (25), so it would require only a relatively small increase in volume to reach an IMP that would compromise the circulation. With normal muscle, however, a 5% increase would appear to have negligible effects on pressure and circulation, and the compartment has to swell to 15% before the pressure increases sufficiently to impair the circulation. The delay in the return to normal of muscle thickness after exercise in the patients implies a continuing venous obstruction following exercise.
Obstructing the blood flow during exercise in healthy control subjects led to significantly higher pain scores that were similar to those of the patients. It is interesting that the pain recovery after exercise was similar in all three groups, as it might be expected that the patients’ pain persists for longer if there was continued increased resistance to blood flow. The finding of increased muscle size after exercise in the patients could explain this as it would allow metabolites to be washed out of the muscle. The limitation to the circulation may only be important when there is a high-energy demand, i.e., during exercise.
In contrast to the force measurements, time course of pain development in the CECS patients was not intermediate between the two control groups but very similar to that reported by the occluded controls throughout exercise and recovery. There was no evidence of the initial lag seen in the fatigue rates of the patients. It is difficult to account for this, but one possibility is that the patient group interprets pain signals arising from working muscles differently.
Although not reported here, the heart rate and blood pressure changes were similar in all three groups, and the patients’ high pain scores were not accompanied by a greater autonomic response. These results are similar to those we have reported previously (4) and could indicate that CECS patients have a tendency to report high pain scores. Alternatively, their autonomic system may have become desensitized by the chronically raised input from muscle chemoreceptors.
The question arises of whether the external venous occlusion used in this study was adequate for its stated purpose. Studies using Doppler ultrasound, performed by an experienced clinical technician, showed that the technique used obstructed tibial flow out of the anterior tibial muscles. Using a thigh cuff inflated to a slightly lower pressure (60–65 mm Hg), Zhang et al. (26) reported a significant increase in intramuscular pressure and a corresponding decrease in muscle blood flow in the anterior tibial muscle. Our results clearly showed a difference between the two exercise conditions in normal subjects, and the increased muscle fatigability, pain, and size with external occlusion demonstrate that it was effective.
It is possible that the patient group may be heterogeneous in etiology, and therefore mechanisms of fatigue and pain could have varied between individuals. Larger controlled studies involving detailed studies of tissue pressure, blood flow, and oxygenation during exercise could resolve this issue.
The results presented here are consistent with the venous outflow being occluded in CECS patients. However, unlike the normal muscle when occluded, the effects develop relatively slowly, and it is interesting to speculate about the mechanism of this delayed effect. If the venous outflow in patients were physically obstructed by, for instance, an abnormally tight fascia, then this would be expected to be immediately effective with the pneumatic cuff on normal subjects. However, this was not the case, and we suggest that the important factor is the accumulation of fluid in the working muscle leading to a progressive increase in pressure and decreased blood flow. In the case of the control subjects with occluded circulation, it is likely that about two thirds of the increased volume was due to venous congestion, and it is notable that this rapidly decreased at the end of the exercise when the cuff was removed. There remains, however, an increased volume of just over 5%, which decreases much more slowly, and this we attribute to increased interstitial fluid. This residual volume was similar to that seen with the patients.
More detailed knowledge about the changes in intramuscular pressure during exercise seems to be central to understanding the cause of CECS. If this increases rapidly at the start of exercise in both normal subjects with occlusion and patients, this would suggest some physical obstruction. Slower increases in the patients would support the idea of a slow build-up of other origin. The invasive nature of intramuscular pressure measurements and technical problems associated with such studies during exercise (15,18,23,24) makes them difficult to carry out. Indirect, noninvasive measurements such as the use of near-infrared spectroscopy for monitoring tissue oxygenation (7,16), photoplethysmography for recording changes in muscle blood flow (26), and measurement of external muscle hardness (11,17) should be revealing.
The Chemical Biological Defense and Human Sciences Domain of the MOD’s Corporate Research Program funded this work.
1. Allen, M. J., and M. R. Barnes. Exercise pain in the lower leg: chronic compartment syndrome and medial tibial syndrome. J. Bone Joint Surg. 68B: 818–823, 1986.
2. Ashton, H. The effect of increased tissue pressure on blood flow. Clin. Orthop. Relat. Res. 113: 15–26, 1975.
3. Bigland-Ritchie, B., C. G. Kukulka, O. C. J. Lippold, and J. J. Woods. Absence of neuromuscular transmission failure in sustained maximal voluntary contractions. J. Physiol. 330: 265–278, 1982.
4. Birtles, D. B., D. Minden, S. Wickes, et al. Muscle function and size during isometric exercise in chronic exertional compartment syndrome of the lower leg. Med. Sci. Sports Exerc. 1900–1906, 2002.
5. Black, K., and D. E. Taylor. Current concepts in the treatment of common compartment syndromes in athletes. Sports Med. 15: 408–418, 1993.
6. Bland, J. M., and D. G. Altman. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1: 307–310, 1986.
7. Breit, G. A., J. H. Gross, D. E. Watenpaugh, B. Chance, and A. R. Hargens. Near-infrared spectroscopy for monitoring of tissue oxygenation of exercising skeletal muscle in a chronic compartment syndrome model. J. Bone Joint Surg. 79-A: 838–843, 1997.
8. Detmer, D. E., K. Sharpe, R. L. Sufit, and F. M. Girdley. Chronic compartment syndrome: diagnosis, management, and outcomes. Am. J. Sports Med. 13: 162–170, 1985.
9. Gershuni, D. H., B. B. Gosink, A. R. Hargens, et al. Ultrasound evaluation of the anterior musculofascial compartment of the leg following exercise. Clin. Orthop. Relat. Res. 167: 185–190, 1982.
10. Hargens, A. R., and S. J. Mubarak. Current concepts in the pathophysiology, evaluation, and diagnosis of compartment syndrome. Hand Clin. 14: 371–383, 1998.
11. Horikawa, M., S. Ebihara, F. Sakai, and M. Akiyama. Non-invasive measurement method for hardness in muscular tissues. Med. Biol. Eng. Comput. 31: 623–627, 1993.
12. Martens, M. A., and J. P. Moeyersoons. Acute and recurrent effort-related compartment syndrome in sports. Sports Med. 9: 62–68, 1990.
13. Martens, M. A., M. Backaert, G. Vermaut, and J. C. Mulier. Chronic leg pain in athletes due to recurrent compartment syndrome. Am. J. Sports Med. 12: 148–151, 1984.
14. Martinson, H., and M. J. Stokes. Measurement of anterior tibial muscle size using real-time ultrasound imaging. Eur. J. Appl. Physiol. 63: 250–254, 1991.
15. Mcdermott, A. G. P., A. E. Marble, R. H. Yabsley, and B. Phillips. Monitoring dynamic anterior compartment pressures during exercise: a new technique using the STIC catheter. Am. J. Sports Med. 10: 83–89, 1982.
16. Mohler, L. R., J. R. Styf, R. A. Pedowitz, A. R. Hargens, and D. H. Gershuni. Intramuscular deoxygenation during exercise in patients who have chronic anterior compartment syndrome of the leg. J. Bone Joint Surg. 79-A: 844–849, 1997.
17. Murayama, M., K. Nosaka, T. Yoneda, and K. Minamitani. Changes in hardness of the human elbow flexor muscles after eccentric exercise. Eur. J. Appl. Physiol. 82: 361–367, 2000.
18. Puranen, J., and A. Alavaikko. Intracompartmental pressure increase on exertion in patients with chronic compartment syndrome in the leg. J. Bone Joint Surg. 63A: 1304–1309, 1981.
19. Qvarfordt, P., J. T. Christenson, B. Eklof, P. Ohlin, and B. Saltin. Intramuscular pressure, muscle blood flow, and skeletal muscle metabolism in chronic anterior tibial compartment syndrome. Clin. Orthop. Relat. Res. 179: 284–289, 1983.
20. Reneman, R. S. The anterior and lateral compartmental syndrome of the leg due to intensive use of muscles. Clin. Orthop. Relat. Res. 113: 69–80, 1975.
21. Rorabeck, C. H., and I. Mcnab. The pathophysiology of the anterior tibial compartment syndrome. Clin. Orthop. Relat. Res. 113: 52–57, 1975.
22. Sjøgaard, G. R., P. Adams, and B. Saltin. Water and ion shifts in skeletal muscle of man with intense dynamic knee-extension. Am. J. Physiol. 248: R190–R196, 1985.
23. Styf, J., L. Körner, and M. Suurkula. Intramuscular pressure and muscle blood flow during exercise in chronic compartment syndrome. J. Bone Joint Surg. 69B: 301–305, 1987.
24. Styf, J. R., and L. M. Körner. Microcapillary infusion technique for measurement of intramuscular pressure during exercise. Clin. Orthop. 207: 253–262, 1986.
25. Turnipseed, W. D., C. Hurschler, and R. Vanderby. The effects of elevated compartment pressure on tibial arteriovenous flow and relationship of mechanical and biochemical characteristics of fascia to genesis of chronic anterior compartment syndrome. J. Vasc. Surg. 21: 810–817, 1995.
26. Zhang, Q., J. Styf, and L.-G. Lindberg. Effects of limb elevation and increased intramuscular pressure on human tibialis anterior muscle blood flow. Eur. J. Appl. Physiol. 85: 567–571, 2001.