 |
 |
 |
|
| home | CME | help | feedback | subscriptions | archive |
|
© 1997 The Journal of Bone and Joint Surgery, Inc. Near-Infrared Spectroscopy for Monitoring of Tissue Oxygenation of Exercising Skeletal Muscle in a Chronic Compartment Syndrome Model*  , MOFFETT FIELD, CALIFORNIAInvestigation performed at National Aeronautics and Space Administration Ames Research Center, Moffett Field
Variations in the levels of muscle hemoglobin and of myoglobin oxygen saturation can be detected non-invasively with near-infrared spectroscopy. This technique could be applied to the diagnosis of chronic compartment syndrome, in which invasive testing has shown increased intramuscular pressure associated with ischemia and pain during exercise. We simulated chronic compartment syndrome in ten healthy subjects (seven men and three women) by applying external compression, through a wide inflatable cuff, to increase the intramuscular pressure in the anterior compartment of the leg. The tissue oxygenation of the tibialis anterior muscle was measured with near-infrared spectroscopy during gradual inflation of the cuff to a pressure of forty millimeters of mercury (5.33 kilopascals) during fourteen minutes of cyclic isokinetic dorsiflexion and plantar flexion of the ankle. The subjects exercised with and without external compression. The data on tissue oxygenation for each subject then were normalized to a scale of 100 per cent (the baseline value, or the value at rest) to 0 per cent (the physiological minimum, or the level of oxygenation achieved by exercise to exhaustion during arterial occlusion of the lower extremity). With external compression, tissue oxygenation declined at a rate of 1.4 ± 0.3 per cent per minute (mean and standard error) during exercise. After an initial decrease at the onset, tissue oxygenation did not decline during exercise without compression. The recovery of tissue oxygenation after exercise was twice as slow with compression (2.5 ± 0.6 minutes) than it was without the use of compression (1.3 ± 0.2 minutes). CLINICAL RELEVANCE: The results of this study demonstrate that near-infrared spectroscopy can detect deoxygenation of skeletal muscle caused by elevated intramuscular pressure during exercise. Thus, this technique may prove to be a useful diagnostic tool for the non-invasive detection of chronic compartment syndrome.
Dual-wave near-infrared spectroscopy is an optical technique that allows non-invasive tracking of variations in the oxygenation of muscle tissue2. Muscle ischemia caused by high intramuscular pressure during exercise occurs in chronic compartment syndrome, usually in the anterior compartment of the leg. With a sufficient increase in pressure, microvascular perfusion is compromised, resulting in ischemia, pain, and sometimes decreased sensibility or neurological dysfunction6,7,18. Chronic compartment syndrome is difficult to diagnose because pain in the lower extremity during exercise is associated with a wide variety of disorders, including periostitis, stress fracture, arterial insufficiency, disease of the central or peripheral nervous system18, and delayed-onset muscular soreness4. Currently, the objective diagnosis of chronic compartment syndrome is made with an invasive method: a catheter is inserted into the muscle to obtain direct measurements of intramuscular pressure. Chronic compartment syndrome may manifest itself in the form of elevated resting pressure before or after exercise or elevated relaxation pressure during cyclic exercise14,20. Like all invasive procedures, however, catheterization of a muscle imposes the risks of infection, bleeding, and discomfort. Currently, there is no non-invasive technique for the objective diagnosis of chronic compartment syndrome. Because the symptoms associated with chronic compartment syndrome are a result of hypoxia of muscle tissues secondary to local ischemia5,16, near-infrared spectroscopy may be an effective means of detecting this syndrome during exercise. Therefore, we simulated chronic compartment syndrome by applying external compression around the leg of healthy subjects. We then used near-infrared spectroscopy to monitor the oxygenation of the tibialis anterior muscle before, during, and after exercise of the leg. We hypothesized that, with external compression, there is a greater decline in the oxygenation of muscle tissue during exercise and a slower recovery after exercise than under normal conditions.
Ten healthy subjects (seven men and three women), with a mean age of twenty-six years (range, eighteen to forty-eight years), gave informed consent to participate in the present study after the protocol had been approved by the Human Research Institutional Review Board. None of the subjects had previously been diagnosed with compartment syndrome or an injury of the leg to be studied. The relative levels of tissue oxygenation and hemoglobin concentration in the tibialis anterior muscle were measured non-invasively with a continuous dual-wave near-infrared spectrophotometer (RunMan; NIM, Philadelphia, Pennsylvania). The principle of near-infrared spectroscopy has been presented previously2. In brief, the technique involves monitoring the absorption of light transmitted through muscle tissue at two distinct wavelengths (760 and 850 nanometers). A change in the oxygenation state of hemoglobin results in opposite changes in the absorption of light at these two wavelengths. By calculating the arithmetic difference between the two absorption signals, the device derives a continuous index of tissue oxygenation that is independent of the over-all state of perfusion of the muscle. The hemoglobin concentration, an indication of in vivo blood volume12, is derived by calculating the sum of the two absorption signals and assuming that the myoglobin concentration remains constant in skeletal muscle. The near-infrared probe (approximately ten by five centimeters) was placed centrally over the tibialis anterior muscle of the dominant leg with an elastic bandage wrapped loosely around the leg. To increase the intramuscular pressure uniformly within the anterior compartment, an eighteen-centimeter-wide inflatable cuff was wrapped around the leg, over the near-infrared probe. A second cuff was placed around the thigh (just proximal to the knee) of the same lower extremity. We applied external compression over the anterior compartment of the leg to increase intramuscular pressures to levels associated with chronic compartment syndrome. Previous data on the transmission of pressure to underlying tissues through an inflatable eighteen-centimeter-wide cuff3,8 indicated that approximately 80 per cent of the inflation pressure of the cuff is transmitted to the intramuscular compartments of the leg. This assumption was the basis of our protocol regarding the application of compression. Next, the subject was placed in a seated bent-knee position in an isokinetic dynamometer (Lido Active; Loredan Biomedical, Davis, California) that was configured for plantar flexion and dorsiflexion of the ankle. The subject continuously monitored the exerted torque, which was displayed as a bar graph on the computer screen of the device. Before each subject participated in the experiment, the maximum torque in dorsiflexion was determined on the basis of the subject's performance of maximum voluntary isometric contractions of three seconds' duration, with the ankle positioned at 20 and 40 degrees from the limit of dorsiflexion. The maximum torque was considered to be the mean of the peak torques that were generated at these two angles. Testing consisted of two minutes of rest, during which baseline data were collected, followed by fourteen minutes of repetitive concentric isokinetic dorsiflexion and plantar flexion of the ankle and sixteen minutes of recovery from exercise. Exercise consisted of full range-of-motion dorsiflexion and plantar flexion, performed at thirty-three cycles per minute in synchrony with a metronome. The subjects were instructed to achieve a torque in dorsiflexion that was equal to 20 per cent of their maximum torque. Plantar flexion resistance was minimum. The actual torque that was exerted by the subject was recorded throughout the duration of exercise. Each subject participated in two different testing protocols, which were carried out on two separate days. On one day, the intramuscular pressure in the tibialis anterior muscle was altered by inflation of the cuff around the leg. Before exercise was begun, the cuff was inflated to ten millimeters of mercury (1.33 kilopascals) to simulate the elevated resting intramuscular pressure that is typically present in patients who have chronic compartment syndrome15. During exercise, the cuff was inflated gradually, at a rate of 2.5 millimeters of mercury (0.33 kilopascal) per minute, from ten to forty millimeters of mercury (1.33 to 5.33 kilopascals). Once exercise had been completed, the cuff was deflated gradually, at the same rate, back to ten millimeters of mercury (1.33 kilopascals), and the data were recorded at this level of pressure during the final two minutes of the experiment. The testing protocol on the other day was identical except for the fact that the cuff was not inflated. The protocol that the subject participated in on the first day of testing was determined randomly, and the subjects were allowed at least two days of rest between tests. On both days of testing, the subjects characterized the degree of pain that they felt during exercise according to a subjective pain index that ranged from 0 (no discomfort) to 10 points (intolerable pain). After completion of the experimental protocol, the physiological minima of tissue oxygenation and hemoglobin concentration were determined. The subjects were instructed to perform repetitive maximum concentric dorsiflexion of the ankle thirty-three times per minute in synchrony with the metronome. After thirty seconds of exercise, blood flow to the leg was occluded by rapid inflation of the cuff on the thigh to approximately 240 millimeters of mercury (thirty-two kilopascals). Exercise was continued until the subject reported exhaustion. The lowest values for tissue oxygenation and hemoglobin concentration indicated by the spectrophotometer during this period of exercise with induced ischemia were recorded as the physiological minima. For analysis of the data, the measurements that were determined with near-infrared spectroscopy for each subject were normalized to a scale that ranged from 0 per cent (the physiological minimum) to 100 per cent (the baseline value, or the value that was determined initially at rest). The values for tissue oxygenation and hemoglobin concentration that were greater than the baseline values are expressed as values greater than 100 per cent. Analysis of variance with repeated measures and post hoc paired t tests were used to determine the significant effects of external compression and duration of exercise and recovery on the normalized values for tissue oxygenation and hemoglobin concentration, as well as on the torque that was exerted. Linear regression analysis was performed to compare the rates of decline of tissue oxygenation during exercise with and without compression. Time constants for recovery of tissue oxygenation to a stable level were estimated by an exponential fit to the data from the end of the fourteen-minute period of exercise to the end of the sixteen-minute period of recovery. The Wilcoxon paired-sample test was used to determine the effect, if any, of compression on the subjective pain index. Linear regression analysis was used to determine whether there was a relationship between the subjective pain index and the level of tissue oxygenation that was recorded just before the completion of exercise with compression. All of the values for tissue oxygenation and hemoglobin concentration, as well as the rates of change in these values, are reported as the mean and the standard error.
During the initial two minutes of rest, the output from both channels of the instrument varied less than ±5 per cent of the normalized scale. During the first two minutes of exercise, the levels of tissue oxygenation with and without compression were similar (82 ± 4 compared with 81 ± 5 per cent of the baseline value) (Fig. 1). During exercise with compression, tissue oxygenation decreased progressively at a rate of 1.4 ± 0.3 per cent per minute, reaching a nadir of 64 ± 3 per cent of the baseline value after fourteen minutes of exercise. In contrast, during exercise without compression, tissue oxygenation remained unchanged (83 ± 4 per cent of the baseline value) after a decrease at the onset of exercise. After the completion of exercise, tissue oxygenation with and without compression recovered to similar levels above the baseline values; however, the recovery was significantly slower (p < 0.05) during testing with compression (2.5 ± 0.6 minutes) than during testing without compression (1.3 ± 0.2 minutes). Four minutes after the cessation of exercise (eighteen minutes after the beginning of exercise), there was no significant difference between the levels of tissue oxygenation with and without compression. The levels remained unchanged, and elevated, during the entire recovery period.
During the first two minutes of exercise with compression, the hemoglobin concentration increased significantly (p < 0.05) to 118 ± 3 per cent of the baseline value and remained elevated at 122 ± 5 per cent of the baseline throughout the entire period of exercise (Fig. 2). In contrast, during exercise without compression, the hemoglobin concentration was unchanged from the baseline value until eight minutes into the period of exercise, at which point it began to increase gradually to a level similar to that observed during exercise with compression. Immediately after the end of exercise, with and without compression, the hemoglobin concentration increased rapidly and remained unchanged throughout the entire period of recovery. However, this increase was greater during recovery with compression (163 ± 9 per cent of the baseline value) than during recovery without compression (140 ± 7 per cent of the baseline value).
The maximum isometric torque that was exerted in dorsiflexion was 25.1 ± 6.3 newton-meters (mean and standard deviation). Over-all, the subjects exerted significantly lower torques (p < 0.0001) when external compression was applied (Fig. 3), despite visual feedback on their performance and coaching to maintain the target level of exertion. The maximum torque was generally unchanged during the period of exercise, except for a significant decline (p < 0.005) from the initial level during the last minute of exercise with compression.
All ten subjects completed testing with no residual untoward effects. The subjects reported a significantly greater subjective pain index (p < 0.003) for exercise with compression (median score, 6 points; range, 1 to 9 points) than for exercise without compression (median score, 1.5 points; range, 0 to 3 points). Furthermore, tissue oxygenation at the completion of exercise with compression correlated inversely with the subjective pain index across the subjects (r2 = 0.506; p < 0.02).
The results demonstrate that near-infrared spectroscopy can detect alterations in the normal pattern of oxygenation in the tibialis anterior muscle that are induced by external compression with a cuff during exercise. When external compression was not applied, the level of tissue oxygenation remained constant throughout the period of exercise (after an initial decrease at the onset). In contrast, gradually increasing pressure with a cuff during exercise caused a progressive decline in tissue oxygenation. This characteristic pattern was consistent across all of the subjects studied and was probably a consequence of ischemia due to compression-induced elevation of intramuscular pressure. Chance and Jöbsis, in 1959, pioneered the use of near-infrared spectroscopy to detect changes in the oxygenation of skeletal muscle. In addition to its use on skeletal muscle, this technique has been applied to the assessment of oxygenation of myocardial10 and cerebral12,17 tissues. A disadvantage of continuous-wave spectrophotometers, such as the device used in the present study, is that they provide only a relative index of tissue oxygenation. Advanced techniques that are based on the detection of photon path lengths, such as pulsed-light spectroscopy2, allow absolute measurement of the oxygen saturation of tissue. However, these techniques are still in the experimental and clinical trial phases. In our study, the limitations of the continuous-wave technique were mostly overcome by normalization of the oxygenation signal to a relative scale that ranged from the baseline value (the value determined at rest) to the physiological minimum (the value measured after the tibialis anterior muscle had been exercised to exhaustion during induced ischemia of the leg). Our compression protocol was designed to simulate the elevated pressure in the anterior compartment that accompanies exercise in patients who have chronic compartment syndrome15. If our model is valid, our results should be similar to the patterns determined with near-infrared spectroscopy in patients who have chronic compartment syndrome. This contention is supported by the findings of Mohler et al., who measured the pressure in the anterior compartment with near-infrared spectroscopy during plantar flexion and dorsiflexion exercise in patients who had chronic compartment syndrome. Those authors also observed a significantly greater degree of deoxygenation during exercise and a significantly slower recovery of oxygenation after exercise in patients who had compartment syndrome than in control subjects (p < 0.05)13. Furthermore, in a third group of patients in that study, who had chronic compartment syndrome-like symptoms but who did not actually have chronic compartment syndrome, the patterns of oxygenation and intramuscular pressure were not substantially different from those in the control group. The time-course of the intramuscular pressures that Mohler et al. observed in patients who had chronic compartment syndrome was qualitatively very similar to that in our model, but the mean resting pressure was of greater magnitude (55.4 millimeters of mercury [7.38 kilopascals]) immediately after exercise than that reported by Pedowitz et al. Although the pain from ischemia associated with chronic compartment syndrome occurs most commonly during exercise, the intramuscular pressure is often increased at rest also. In a previous investigation15, the resting pressure in the tibialis anterior muscle was found to be 15.5 ± 6.8 millimeters of mercury (2.07 ± 0.91 kilopascals) in patients who had been diagnosed with chronic compartment syndrome and 7.9 ± 3.4 millimeters of mercury (1.05 ± 0.45 kilopascals) in healthy volunteers. In the present study, we initially inflated the cuff to a pressure of ten millimeters of mercury (1.33 kilopascals) to simulate this elevation in resting pressure, with the assumption that 80 per cent of the cuff pressure would be transmitted to the underlying muscle3,8. In the study by Pedowitz et al., a pressure of 44.3 ± 16.5 millimeters of mercury (5.91 ± 2.20 kilopascals) was found one minute after exercise to exhaustion in patients who had chronic compartment syndrome, compared with 12.2 ± 8.0 millimeters of mercury (1.63 ± 1.07 kilopascals) in patients who did not have chronic compartment syndrome. On the basis of previous studies on the transmission of pressure3,8, we simulated this pressure differential of thirty-two millimeters of mercury (4.27 kilopascals) by inflating the cuff to a final pressure of forty millimeters of mercury (5.33 kilopascals), although this protocol was not verified by direct measurement of the intramuscular pressure. The mean compartmental pressure immediately after exercise in patients who had chronic compartment syndrome was substantially greater in the study by Mohler et al. than in the study by Pedowitz et al., on whose results we modeled our testing protocol. This may explain the more rapid deoxygenation during exercise and the slower recovery after exercise in the study by Mohler et al. compared with those values in the present study. The progressive decrease in tissue oxygenation that we observed during exercise with external compression is consistent with the findings of previous authors who used direct invasive methods to assess the partial pressure of oxygen in muscle7,16 and those who used phosphorus nuclear magnetic resonance spectroscopy9. Those studies demonstrated decreased levels of intramuscular oxygen and deranged cellular metabolism, respectively, with increased compartmental pressure. In the present study, the decrease in tissue oxygenation that was recorded with near-infrared spectroscopy suggested that, by the end of the exercise protocol, pain and exhaustion due to ischemia were imminent. Indeed, the subjects reported a significantly greater level of discomfort during exercise with compression than during exercise without compression (p < 0.003). In all subjects, this discomfort was correlated with the level of deoxygenation that was recorded at the end of exercise with compression (r2 = 0.506; p < 0.02). Furthermore, the significant decline (p < 0.005) in the amount of torque that was exerted during the last minute of exercise with compression indicates ischemia-induced muscle fatigue. In patients who have chronic compartment syndrome, pain that develops during exercise subsides rapidly after activity has stopped. However, the abnormally high intramuscular pressure tends to persist for several minutes. Pedowitz et al. observed that five minutes after exercise the pressure in the tibialis anterior muscle was 32.5 ± 13.4 millimeters of mercury (4.33 ± 1.79 kilopascals) in patients who had chronic compartment syndrome and 8.9 ± 4.9 millimeters of mercury (1.19 ± 0.65 kilopascals) in control subjects. In the present study, we modeled this persistence of compartmental tamponade by gradually reducing the pressure exerted by the cuff during recovery at the same rate (2.5 millimeters of mercury [0.33 kilopascal] per minute) as it had been increased during exercise. Our hypothesis had been that the recovery of tissue oxygenation would be slower after exercise with external compression than after exercise without compression. The data for tissue oxygenation during the time-interval from the end of the fourteen-minute period of exercise to the end of the sixteen-minute period of recovery were fit with an exponential decay model to determine time constants for recovery. In agreement with our hypothesis, the recovery of tissue oxygenation was significantly slower (p < 0.05) after exercise with compression. Despite the disparity in the time for recovery of tissue oxygenation, there was no significant difference between the levels of tissue oxygenation during testing with and without compression by four minutes after the completion of exercise (eighteen minutes after the initiation of exercise) (Fig. 1). During that same time, however, the hemoglobin concentration (the index of blood volume) was significantly greater (p < 0.05) during testing with compression (Fig. 2), which indicates enhanced hyperemia after exercise during ischemia. The similarity in the levels of tissue oxygenation during testing with and without compression suggests a higher rate of cellular uptake of oxygen during recovery from exercise with compression. The present study demonstrates that near-infrared spectroscopy is sensitive enough to detect deoxygenation of muscle caused by elevated intramuscular pressure, accomplished by external compression, in skeletal muscle during exercise. In conjunction with the results of Mohler et al., our data indicate that this technique is a promising non-invasive diagnostic tool for the detection of chronic compartment syndrome.
NOTE: The authors gratefully acknowledge Jorma R. Styf, M.D., Ph.D., for helpful discussions on the clinical relevance and application of this technology; Charles E. Wade, Ph.D., and Sara B. Arnaud, M.D., for their thorough review of the manuscript; Richard E. Ballard, Gita Murthy, and Karen J. Hutchinson, for technical assistance; and NIM, for the loan of the near-infrared equipment.
*One or more of the authors has received or will receive benefits for personal or professional use from a commercial party related directly or indirectly to the subject of this article. Funds were received in total or partial support of the research or clinical study presented in this article. The funding sources were National Aeronautics and Space Administration Grants 199-14-12-04 and 199-26-12-34, National Institutes of Health Grant HL44125-05, and a postdoctoral fellowship (G. A. B.) from the National Research Council.
  CiteULike  Connotea  Del.icio.us  Facebook  Technorati  Twitter What's this?
This article has been cited by other articles:
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Top
Abstract
Introduction
| home | CME | help | feedback | subscriptions | archive | site index |
|
Stanford University Libraries' HighWire Press (TM) assists in the publication of JBJS Online. |