The tibia is the most commonly fractured long bone in adults1. A potentially devastating complication associated with tibial fractures is the development of compartment syndrome. The prevalence of compartment syndrome has been reported to be as high as 10% in closed tibial fractures2, and as high as 20% in the intensive care setting3. Near-infrared spectroscopy measures the percentage of hemoglobin saturated with oxygen in the microcirculation of tissue within approximately 2 to 3 cm below the skin4-7. This technology has been used to examine both acute and chronic compartment syndromes of the lower leg in prior studies8-12. However, no data have been published regarding near-infrared spectroscopy values in patients without compartment syndrome.
Near-infrared spectroscopy has been studied, validated, and approved by the U.S. Food and Drug Administration in the anesthesia setting to monitor cerebral oxygenation4,13. Light in the near-infrared range (600 to 1000 nm) is capable of penetrating through skin, soft tissue, and bone. The majority of light absorption is based on the relative concentrations of oxygenated and deoxygenated hemoglobin in the microcirculation. Since large vessels or hematomas absorb the light completely, the only light collected by the sensors is light that is capable of passing through the microcirculation4. By using the Beer-Lambert law and two separate wavelengths of near-infrared light, the concentration of both oxygenated and deoxygenated hemoglobin can be estimated14. The depth of tissue penetration (or the location of tissue oxygenation measurement) is directly proportional to the distance or separation between the light source and the light receptor5,6.
Cerebral oximetry with use of near-infrared spectroscopy was shown to reflect a weighted average of approximately 3 to 1 of jugular (venous) to carotid (arterial) oxygen saturation5,15,16. Since cerebral blood by volume is more venous than arterial, a venous weighted average is consistent with the blood oxygenation as a whole in the brain17. When the venous oxygenation status is used to monitor perfusion, the relationship between the oxygen supply and consumption can be monitored. If tissue perfusion does not match oxygen consumption, more oxygen will be extracted from the arterial blood, resulting in lower venous oxygen saturation. Therefore, venous weighted oxygen saturation reflected in near-infrared spectroscopy measurements can be used as a surrogate for tissue perfusion.
Multiple studies have validated cerebral oximetry. The near-infrared spectroscopy responsiveness and changes in jugular oxygen saturation in normal and hypoxic volunteers were shown to be highly correlated4,5,15-19. Additional studies have shown minimal effects of subcutaneous blood flow when monitoring cerebral oxygenation during carotid endarterectomy. Only a 2% change in tissue oxygenation was seen when the external carotid artery, the arterial blood supply to the subcutaneous tissue of the forehead, was clamped off during the procedure15,16,20,21. Through multiple investigations, near-infrared spectroscopy has been shown to offer a reliable, noninvasive, continual, and real-time means of monitoring tissue oxygenation in the brain and in muscle18,19,22.
The purpose of this study was to describe the expected alteration of normal tissue oxygenation in the lower leg in the setting of acute trauma without compartment syndrome. Secondarily, the utility of using the contralateral leg as a control measurement was examined. Without knowledge of what happens in trauma patients without compartment syndrome, a deviation from these norms cannot be identified in patients who potentially have a compartment syndrome.
Materials and Methods
The study group consisted of twenty-six consecutive patients with an acute unilateral tibial fracture, and the control group consisted of twenty-five uninjured volunteer subjects. All participants were recruited at a level-I trauma center between February 25, 2007, and July 1, 2007, after approval from the institutional review board.
For the trauma group, the inclusion criteria consisted of unilateral tibial fractures, including proximal intra-articular (plateau), tibial shaft, and distal intra-articular (pilon) fractures as well as open fractures. Exclusion criteria for the injured group included bilateral lower extremity injury, a previous diagnosis of pulmonary or vascular disease, or an inability and/or unwillingness to provide informed consent. Injuries occurring more than sixty hours prior to measurement were excluded since complications associated with acute injury are less likely after this time.
Twenty-six consecutive trauma patients who met the inclusion criteria were studied. No patient who was asked to participate declined to participate. Four patients were excluded from participation on the basis of a possible or verified compartment syndrome at the time of the initial evaluation.
Uninjured control subjects were recruited from the available pool of family members of clinic patients and clinical staff who were willing to volunteer for the study. Exclusion criteria for the uninjured control group included a previous diagnosis of pulmonary or vascular disease, acute injury of any type, or an inability or unwillingness to provide informed consent. Controls were selected to represent a demographically diverse population of healthy nonelderly adults.
For each participant of both the injured and control groups, near-infrared spectroscopy measurements were obtained, with use of the INVOS cerebral oximeter (model 41000; Somanetics, Troy, Michigan) from both the right and left legs at the mid-tibial level for each of the four muscle compartments: anterior, lateral, deep posterior, and superficial posterior. The device consists of a disposable adhesive sensor pad with wires leading to a processor and a display monitor. Near-infrared spectroscopy readings for tissue oxygenation were displayed in the form of a percentage representing the proportion of hemoglobin saturated with oxygen; therefore, a higher reading or percentage is indicative of higher tissue oxygenation. On the basis of the set distance between the light source and the light sensors in the INVOS pad, the depth of tissue monitoring is between 2 and 3 cm deep to the surface of the skin5. The INVOS sensor uses two separate measurement depths, and the shallow values are subtracted from the deeper readings in order to isolate the perfusion of the tissue at the deeper level5.
Readings were obtained by placing the sensor pad over the middle one-third of the tibia for all four compartments. The sensor was applied to the leg for approximately thirty to sixty seconds to obtain a stable reading. The near-infrared spectroscopy device cycles every six seconds to generate a new reading. A stable reading was defined as four consecutive cycles with the same value. A stable reading was obtained within sixty seconds in all legs.
In all injured patients, the tibial fracture was provisionally reduced with traction in order to restore the relative anatomical position. Traction was not applied during the readings in order to minimize any increase in intracompartmental pressure23,24. All measurements were made with the heel placed on a rolled towel to remove the pressure caused by the weight of the leg. The foot was maintained in a neutral position. All patients were in a reclined supine position at the time of measurement. Readings were obtained once the patient was hemodynamically stable. Calibration of the device is performed at the time of manufacture and is retained on a microchip in the sensor and monitor. The calibration algorithm is based on the Beer-Lambert law that is modified for spatial resolution (looking at the slope of extinction at progressively deeper penetration depths). Since the blood being measured is a mixture of both oxygenated and deoxygenated blood, the values fall between the two extinction curves for deoxygenated hemoglobin (0%) and oxygenated hemoglobin (100%)5. The system does not require additional calibration at the time of use25.
The anterior compartment was accessed by placing the pad lateral to the anterior tibial ridge. The lateral compartment was measured over the anterior aspect of the fibula on the lateral aspect of the leg. The superficial posterior compartment measurement was made by placing the pad on the posterior aspect of the leg. The deep posterior compartment was measured by placing the sensor just posterior to the medial aspect of the tibia over the flexor digitorum longus. The medial aspect of the tibia was palpated until the posterior border was identified, and the pad was placed so that the light would be directed just posterior to the tibia. By placing the sensor at this point, it is possible to obtain a deep posterior measurement, while limiting the potential interference from the superficial posterior compartment (Fig. 1).
The mid-tibial level was selected for measurements in order to compare uninjured and injured subjects. Since no data exist on near-infrared spectroscopy values at different locations in the leg, the goal was to eliminate potential confounding factors by maintaining a consistent location on the leg for measurements. Additionally, the mid-tibial level maximized the muscle cross-sectional area while limiting superficial posterior compartment overlap of the deep compartment.
For a few limbs with a tibial fracture, the technique for obtaining near-infrared spectroscopy measurements required modification. The near-infrared spectroscopy values are not obtainable over hematomas, since the light is completely absorbed by the collection of blood4. If a reading at the mid-tibial level was unobtainable, a stable reading was obtained as close to the middle of the leg as possible by moving the sensor either proximally or distally along the axis of the compartment. In the case of open wounds, the skin was manually reapproximated at the time of fracture reduction in the emergency department in order to obtain a reading over intact skin. This process was performed to ensure that any effects of the skin were consistent. All measurements were obtained within the middle one-third of the tibia despite these alterations.
The analysis was designed to compare the injured legs of twenty-six patients who had an acute tibial fracture and the uninjured legs (randomly selected) of twenty-five control patients, with the contralateral, uninjured leg of each patient used as an internal control for baseline oxygen saturation. The near-infrared spectroscopy values of injured, uninjured, and contralateral (all uninjured) legs were tabulated as means, standard deviations, and ranges. The correlation between alternate legs of uninjured subjects was described with use of the Pearson correlation coefficient. Differences in crude baseline values between black and white subjects were tested with use of the two-sample t test. The effect of trauma on the near-infrared spectroscopy values was evaluated by repeated-measures analysis with use of a generalized linear mixed model, which controlled for the near-infrared spectroscopy value of the contralateral limb. In addition to summarizing the average differences between fractured and uninjured legs across the four compartments, the repeated-measures model allowed us to test directly whether average near-infrared spectroscopy values differed among any of the compartments, as well as whether the amount of difference in near-infrared spectroscopy values associated with the presence compared with the absence of fracture differed among any of the compartments. Because skin pigmentation could affect the near-infrared spectroscopy measurement, race was included as a covariate. Linear regression was used to obtain R2 values for the prediction of near-infrared spectroscopy values within each compartment individually. The R2 value represents the proportion of the variance in an outcome variable (i.e., the near-infrared spectroscopy value) that is explained by predictor variables (i.e., fracture status and near-infrared spectroscopy value in the contralateral leg). We used R2 in this study to quantify the utility of the near-infrared spectroscopy value in the contralateral limb as an internal control. Alpha was set at 0.05 (two-sided) for all statistical tests.
A prestudy power analysis had determined that twenty-four study patients and twenty-four control subjects would provide 80% power to detect a bivariate difference at a two-sided significance level of 0.05 if the true difference between the mean near-infrared spectroscopy values of the injured and uninjured patients was at least 5 percentage points. The analysis assumed a normally distributed response variable with a standard deviation of 6.1 percentage points, which was based on a pilot study of uninjured subjects.
Source of Funding
There was no external funding source. However, Somanetics donated the near-infrared spectroscopy equipment for the purpose of the study.
The demographic characteristics of the participants are shown in Table I. For the injured group, the lower extremity trauma consisted of a plateau fracture (four patients), tibial shaft fracture (seventeen patients), or pilon fracture (five patients). There were seven open tibial fractures. One of them was classified as a Gustilo grade I, while three were grade-II and three were grade-IIIA injuries26,27. The average time between injury and measurement was 16 ± 12 hours (range, two to fifty-two hours). A motor vehicle accident was responsible for the injury in sixteen patients, while a fall accounted for seven injuries and a pedestrian-motor vehicle accident accounted for three injuries. At the time of measurement, all patients were hemodynamically stable with an average blood pressure of 133/72 mm Hg (Table I).
The mean values for the anterior, lateral, deep posterior, and superficial posterior compartments were 69%, 70%, 74%, and 70%, respectively, for the injured legs and 54%, 55%, 60%, and 57% for the uninjured control group (Tables II and III). For the uninjured volunteers, the near-infrared spectroscopy values were highly correlated between the control limbs and the contralateral limbs in all compartments (Pearson product moment correlation, r = 0.888, 0.866, 0.879, and 0.905, respectively; p < 0.0001 for all). The black patients had substantially lower mean raw near-infrared spectroscopy values than the white patients (Table IV).
The repeated-measures model revealed that near-infrared spectroscopy values were an estimated 15.4 percentage points (95% confidence interval, 12.2 to 18.6 percentage points) higher among the injured legs than among the uninjured legs across the four compartments, controlling for the baseline value of the contralateral leg and race (p < 0.0001). The model results are summarized in Table V. The alteration of near-infrared spectroscopy associated with the presence of a fracture did not differ significantly among any of the four compartments (p = 0.9253 for effect of compartment × fracture interaction); however, the values themselves were not equal across the four compartments (p = 0.0004 for effect of compartment) (Table V). Post hoc contrast tests revealed that the deep posterior compartment differed modestly from each of the other three compartments, while the anterior, lateral, and superficial posterior compartments were not significantly different from one another. The largest of those differences was between the deep posterior and anterior compartments (estimated difference, 2.6 percentage points; 95% confidence interval, 0.7 to 4.6 percentage points). These modest differences of the deep posterior from other compartments are also demonstrated in Tables II and III. The near-infrared spectroscopy value of the contralateral limb was a highly significant covariate (p < 0.0001), and the near-infrared spectroscopy values were furthermore lower among blacks compared with whites in the controlled model (p = 0.0173) (Table V).
The presence compared with the absence of a fracture, together with the near-infrared spectroscopy value of the contralateral limb, explained an extremely high proportion (approximately 74% to 80%) of the variance of the near-infrared spectroscopy values that was observed among the fifty-one legs (R2 = 0.74, 0.75, 0.79, and 0.80 for the anterior, lateral, deep posterior, and superficial posterior compartments, respectively), whereas fracture status alone (without the use of the contralateral limb as a control) explained a substantially smaller proportion of only 30% to 47% (R2 = 0.47, 0.41, 0.37, and 0.30 for the respective compartments).
There were no adverse events due to the use of the investigational near-infrared spectroscopy device among any subjects.
The body's response to injury is to increase blood flow to the site of trauma. In 1929, Bradburn and Blalock showed a decrease in the arteriovenous oxygen difference in blood from an injured limb28. In 1970, Lewis and Lim showed that the increased blood flow after trauma was through capillary beds rather than shunting through larger vessels29,30. Imms et al., in 1975, also described a lasting hyperemic effect in soft tissue surrounding a healing fracture of the tibia31. In other words, there is a hyperemic response that causes an increase in oxygen saturation in the venous capillary system in response to injury. Since near-infrared spectroscopy measurements reflect a weighted venous arterial average in the capillary bed of soft tissues, the resultant increase in near-infrared spectroscopy values, a measure of microcirculation oxygenation, is consistent with these previous reports5.
In a study performed on anesthetized dogs, Sandegard and Zachrisson showed with angiograms that vasodilation occurs within thirty seconds in response to trauma in an injured lower extremity32-34. There was a decrease in vascular resistance in the injured extremity with a concomitant and inverse increase in resistance in the uninjured leg32,33. This response was sustained for several days to weeks after injury35. These observations also confirm the hyperemic findings in the injured limbs of this study.
The results from this study show that lower extremity trauma causes a predictable increase in near-infrared spectroscopy values in the absence of compartment syndrome. With all factors considered, a fracture is associated with an average 15.4 percentage point increase in near-infrared spectroscopy values compared with the nonfractured state. This represents an increase of roughly 25% to 30% of baseline values. This hyperemic effect is consistent across all four compartments.
The corresponding compartment of the contralateral leg offered strong utility as an internal control value when evaluating the response to an injury (i.e., the proportion of variation in the near-infrared spectroscopy values explained by our linear regression models increased dramatically with the inclusion of the contralateral limb value—from between 30% and 47% before to between 74% and 80% after the inclusion). The injured extremity showed a hyperemic response in all compartments when the injured extremity was compared with both the uninjured, contralateral side and the uninjured limbs in the study group. The near-infrared spectroscopy values of the tibial compartments appear highly variable among individuals, but they were highly correlated between limbs of the same individual. The uninjured side in the trauma group had mean values (55% to 57%) (Table III) that were similar to those of the uninjured study group (55% to 59%) (Table II), further suggesting that the uninjured side of a trauma patient should be an appropriate control. In our models, it appeared that predictive strength was shared approximately equally by fracture status and near-infrared spectroscopy value of the contralateral limb.
Skin pigmentation does have an effect upon near-infrared spectroscopy values. Black or darker pigmented subjects had attenuated raw near-infrared spectroscopy values, differing by approximately 9 percentage points on the average. The decreased near-infrared spectroscopy values in darker pigmented subjects likely occur because skin pigments absorb more light. This trend is consistent with the finding by Wassenaar and Van den Brand, who examined skin pigmentation in relationship to signal loss in human volunteers with use of a chronic exertional compartment syndrome model36. They concluded that melanin played a significant role in the near-infrared spectroscopy signal and light absorption (p = 0.012).
Attempts have been made to apply near-infrared spectroscopy in the setting of acute compartment syndrome. An initial animal study showed that near-infrared spectroscopy was inversely related to intracompartmental pressures with use of an infusion compartment syndrome model in pigs10. A follow-up study demonstrated the responsiveness of near-infrared spectroscopy in the setting of both hypotension and hypoxemia8. In a calf compression model with use of human volunteers, near-infrared spectroscopy was shown to be more sensitive to the ischemic condition measured by nerve conduction studies than when measured by perfusion pressure37. Last, in established compartment syndromes, near-infrared spectroscopy values were lower compared with near-infrared spectroscopy values in trauma patients without compartment syndrome11. In order to interpret near-infrared spectroscopy values in the trauma setting, normal values for both an uninjured and an injured patient without compartment syndrome must be established.
The results of this study, as well as previous studies, demonstrate that a normal response to soft tissue and osseous trauma is hyperemia and increased perfusion compared with both the uninjured, contralateral leg as well as an uninjured control group. If there is an absence of hyperemia, the clinician should be concerned about impaired oxygenation in the injured leg. The effects of peripheral vascular disease and diabetes are unknown with regard to this response, and additional investigation regarding the effects is required.
These findings have implications when considering the pathophysiology of compartment syndrome. Prior to intracompartmental pressures causing ischemia, these results suggest that there is actually a hyperperfusion period due to a decrease in vascular resistance in response to trauma. In previously described models for compartment syndrome, in which perfusion and intracompartmental pressures were correlated, the response associated with traumatic events has been ignored8,10,38,39.
There are limitations to this study. As was shown by Heckman et al., the pressure within a compartment can vary on the basis of the distance from the fracture40. Since measurements were obtained only at the mid-tibial level, the possibility of different pressures and perfusion along the compartment may not have been appreciated. Additionally, the near-infrared spectroscopy measurements were obtained for only a short period of time, roughly sixty seconds, which limits the information that can be obtained concerning a continuous monitoring system. The hyperemia associated with trauma was consistent despite a wide range of injury patterns and time from injury in this study. Disorders that affect peripheral microcirculation, such as diabetes and peripheral vascular disease, were not examined in this patient cohort. The ability to fully elucidate the effect of skin pigmentation was limited by the sample size and the lack of a quantitative measure of pigmentation. While the deep posterior compartment was measured at the posteromedial border of the tibia, which limits the overlap of the superficial posterior compartment, the proximal half of the tibia typically has some overlap of the superficial posterior compartment. There is evidence to show that the INVOS device does isolate deep-tissue values as it was designed, but there is no so-called gold standard for measurements in the leg5,16,18,19. Therefore, the ability of the device to measure the oxygenation of the deep posterior compartment is uncertain.
The findings from this study demonstrate that a predictable hyperemic response occurs in the acute setting of lower extremity trauma. This response is reproducibly detected by near-infrared spectroscopy monitoring. Additionally, the uninjured, contralateral leg appears to provide a highly effective internal control to interpret the findings in the injured leg. Additional studies are required to evaluate the utility of near-infrared spectroscopy in the evaluation of lower extremity trauma and potential compartment syndromes.
Disclosure: In support of their research for or preparation of this work, one or more of the authors received, in any one year, outside funding or grants of less than $10,000 from Somanetics. Neither they nor a member of their immediate families received payments or other benefits or a commitment or agreement to provide such benefits from a commercial entity. No commercial entity paid or directed, or agreed to pay or direct, any benefits to any research fund, foundation, division, center, clinical practice, or other charitable or nonprofit organization with which the authors, or a member of their immediate families, are affiliated or associated.
Investigation performed at Grady Memorial Hospital and Emory University, Atlanta, Georgia
1. Bhandari M, Guyatt GH, Swiontkowski MF, Tornetta P 3rd, Hanson B, Weaver B, Sprague S, Schemitsch EH. Surgeons' preferences for the operative treatment of fractures of the tibial shaft. An international survey. J Bone Joint Surg Am. 2001;83:1746-52.
2. Court-Brown CM, Will E, Christie J, McQueen MM. Reamed or unreamed nailing for closed tibial fractures. A prospective study in Tscherne C1 fractures. J Bone Joint Surg Br. 1996;78:580-3.
3. Kosir R, Moore FA, Selby JH, Cocanour CS, Kozar RA, Gonzalez EA, Todd SR. Acute lower extremity compartment syndrome (ALECS) screening protocol in critically ill trauma patients. J Trauma. 2007;63:268-75.
4. Mancini DM, Bolinger L, Li H, Kendrick K, Chance B, Wilson JR. Validation of near-infrared spectroscopy in humans. J Appl Physiol. 1994;77:2740-7.
5. Kim MB, Ward DS, Cartwright CR, Kolano J, Chlebowski S, Henson LC. Estimation of jugular venous O2 saturation from cerebral oximetry or arterial O2 saturation during isocapnic hypoxia. J Clin Monit Comput. 2000;16:191-9.
6. Arimoto H, Egawa M, Yamada Y. Depth profile of diffuse reflectance near-infrared spectroscopy for measurement of water content in skin. Skin Res Technol. 2005;11:27-35.
7. Meyer RS, White KK, Smith JM, Groppo ER, Mubarak SJ, Hargens AR. Intramuscular and blood pressures in legs positioned in the hemilithotomy position: clarification of risk factors for well-leg acute compartment syndrome. J Bone Joint Surg Am. 2002;84:1829-35.
8. Arbabi S, Brundage SI, Gentilello LM. Near-infrared spectroscopy: a potential method for continuous, transcutaneous monitoring for compartmental syndrome in critically injured patients. J Trauma. 1999;47:829-33.
9. Breit GA, Gross JH, Watenpaugh DE, Chance B, Hargens AR. Near-infrared spectroscopy for monitoring of tissue oxygenation of exercising skeletal muscle in a chronic compartment syndrome model. J Bone Joint Surg Am. 1997;79:838-43.
10. Garr JL, Gentilello LM, Cole PA, Mock CN, Matsen FA 3rd. Monitoring for compartmental syndrome using near-infrared spectroscopy: a noninvasive, continuous, transcutaneous monitoring technique. J Trauma. 1999;46:613-8.
11. Giannotti G, Cohn SM, Brown M, Varela JE, McKenney MG, Wiseberg JA. Utility of near-infrared spectroscopy in the diagnosis of lower extremity compartment syndrome. J Trauma. 2000;48:396-401.
12. Mohler LR, Styf JR, Pedowitz RA, Hargens AR, Gershuni DH. Intramuscular deoxygenation during exercise in patients who have chronic anterior compartment syndrome of the leg. J Bone Joint Surg Am. 1997;79:844-9.
13. Murkin JM, Adams SJ, Novick RJ, Quantz M, Bainbridge D, Iglesias I, Cleland A, Schaefer B, Irwin B, Fox S. Monitoring brain oxygen saturation during coronary bypass surgery: a randomized, prospective study. Anesth Analg. 2007;104:51-8.
14. Fadel PJ, Keller DM, Watanabe H, Raven PB, Thomas GD. Noninvasive assessment of sympathetic vasoconstriction in human and rodent skeletal muscle using near-infrared spectroscopy and Doppler ultrasound. J Appl Physiol. 2004;96:1323-30.
15. Samra SK, Dy EA, Welch K, Dorje P, Zelenock GB, Stanley JC. Evaluation of a cerebral oximeter as a monitor of cerebral ischemia during carotid endarterectomy. Anesthesiology. 2000;93:964-70.
16. Samra SK, Stanley JC, Zelenock GB, Dorje P. An assessment of contributions made by extracranial tissues during cerebral oximetry. J Neurosurg Anesthesiol. 1999;11:1-5.
17. Henson LC, Calalang C, Temp JA, Ward DS. Accuracy of a cerebral oximeter in healthy volunteers under conditions of isocapnic hypoxia. Anesthesiology. 1998;88:58-65.
18. Boushel R, Pott F, Madsen P, Rådegran G, Nowak M, Quistorff B, Secher N. Muscle metabolism from near infrared spectroscopy during rhythmic handgrip in humans. Eur J Appl Physiol Occup Physiol. 1998;79:41-8.
19. Kaufman J, Almodovar MC, Zuk J, Friesen RH. Correlation of abdominal site near-infrared spectroscopy with gastric tonometry in infants following surgery for congenital heart disease. Pediatr Crit Care Med. 2008;9:62-8.
20. Duncan LA, Ruckley CV, Wildsmith JA. Cerebral oximetry: a useful monitor during carotid artery surgery. Anaesthesia. 1995;50:1041-5.
21. Grubhofer G, Lassnigg A, Manlik F, Marx E, Trubel W, Hiesmayr M. The contribution of extracranial blood oxygenation on near-infrared spectroscopy during carotid thrombendarterectomy. Anaesthesia. 1997;52:116-20.
22. Austin KG, Daigle KA, Patterson P, Cowman J, Chelland S, Haymes EM. Reliability of near-infrared spectroscopy for determining muscle oxygen saturation during exercise. Res Q Exerc Sport. 2005;76:440-9.
23. Tornetta P 3rd, French BG. Compartment pressures during nonreamed tibial nailing without traction. J Orthop Trauma. 1997;11:24-7.
24. Tornetta P 3rd, Templeman D. Compartment syndrome associated with tibial fracture. Instr Course Lect. 1997;46:303-8.
25. Levy WJ, Carpenter J, Fairman RM, Golden MA, Zager E. The calibration and validation of a phase-modulated near-infrared cerebral oximeter. J Clin Monit Comput. 1999;15:103-8.
26. Gustilo RB, Anderson JT. Prevention of infection in the treatment of one thousand and twenty-five open fractures of long bones: retrospective and prospective analyses. J Bone Joint Surg Am. 1976;58:453-8.
27. Gustilo RB, Mendoza RM, Williams DN. Problems in the management of type III (severe) open fractures: a new classification of type III open fractures. J Trauma. 1984;24:742-6.
28. Bradburn HB, Blalock A. The relationship of changes in blood-flow through an extremity to changes in temperature of tissue, differences in oxygen content of the arterial and venous blood, and cardiac output. Am J Physiol. 1929;91:115-22.
29. Lewis DH, Lim RC Jr. Studies on the circulatory pathophysiology of trauma. II. Effect of acute soft tissue injury on the passage of macroaggregated albumin (131-I) particles through the hindleg of the dog. Acta Orthop Scand. 1970;41:37-43.
30. Lewis DH, Lim RC Jr. Studies on the circulatory pathophysiology of trauma. I. Effect of acute soft tissue injury on nutritional and non-nutritional shunt flow through the hindleg of the dog. Acta Orthop Scand. 1970;41:17-36.
31. Imms FJ, Prestidge SP, Thornton C. Proceedings: blood flows in the lower leg during recovery from fractures of the tibia and fibula. J Physiol. 1975;251:22P.
32. Sandegard J, Zachrisson BE. Angiography and hemodynamic measurements in extensive soft tissue trauma to the extremity. Acta Radiol Diagn (Stockh). 1975;16:279-96.
33. Sandegard J, Zachrisson BE. Circulatory disturbances after experimental fracture. Acta Radiol Diagn (Stockh). 1975;16:181-92.
34. Sandegard J. Vasodilatation in extremity trauma. Immediate hemodynamic changes in the dog hind leg. Acta Chir Scand Suppl. 1974;447:1-32.
35. Sandegard J. Muscle blood flow in patients with ankle fracture. J Trauma. 1974;14:873-9.
36. Wassenaar EB, Van den Brand JG. Reliability of near-infrared spectroscopy in people with dark skin pigmentation. J Clin Monit Comput. 2005;19:195-9.
37. Gentilello LM, Sanzone A, Wang L, Liu PY, Robinson L. Near-infrared spectroscopy versus compartment pressure for the diagnosis of lower extremity compartmental syndrome using electromyography-determined measurements of neuromuscular function. J Trauma. 2001;51:1-9.
38. Clayton JM, Hayes AC, Barnes RW. Tissue pressure and perfusion in the compartment syndrome. J Surg Res. 1977;22:333-9.
39. Dahn I, Lassen NA, Westling H. Blood flow in human muscles during external pressure or venous stasis. Clin Sci. 1967;32:467-73.
40. Heckman MM, Whitesides TE Jr, Grewe SR, Rooks MD. Compartment pressure in association with closed tibial fractures. The relationship between tissue pressure, compartment, and the distance from the site of the fracture. J Bone Joint Surg Am. 1994;76:1285-92.