Over the course of this project we improved our understanding of how light interacts with muscle covered by thick layers of fat. We developed a mathematical model of spectra for human tissue which takes into account the skin, fat and muscle layers. The model was used, in conjunction with tissue mimicking phantoms, to better understand the influence that fat thickness and its absorbance and scattering properties have on the absorption spectrum used to calculate muscle pH and PO2. We created a novel tissue mimicking phantom based on a near infrared dye with absorption properties similar to deoxyhemoglobin. Measurement on these phantoms was used to demonstrate prediction capabilities of the Monte Carlo model used to study light interaction with multi-layer tissue structures. Using these phantom materials we developed and demonstrated new methods for correcting tissue absorption spectra for skin pigment and fat using a novel 2-source fiber optic sensor and mathematics to allow us to isolate the spectrum of the muscle from the spectrum of the skin and fat.

The phantoms and Monte Carlo model helped us identify key sensor design issues to assure optical depth penetration through fat on the forearm and the thigh. One system was optimized for thinner fat layers and was made available to NASA-JSC for handgrip studies and measurements on the calf during treadmill exercise. We redesigned this sensor and the accompanying monitor to achieve good muscle spectra through thick layers of fat overlying thigh muscle. One of these systems was delivered to NASA-JSC for thigh measurements during treadmill studies.

A new method for calculating muscle oxygen saturation (SmO2) was developed and its precision was determined as part of a handgrip study at NASA-JSC. We also work with NASA-JSC to demonstrate that we could use the pH equation to noninvasively determine hydrogen ion threshold as a surrogate for lactate threshold during cycle ergometry.

Another goal of our project was to develop noninvasive methods to measure absolute values for key parameters so they can be used as part of a smart medical system to help diagnose and guide treatment for critical injuries. To do this we needed to establish normal values for new parameters and be able to identify values of these new parameters that indicate when someone is sick or getting better. Working with the US Army we used lower body negative pressure (LBNP) as a model to simulate the early stages of hemorrhagic shock or internal bleeding. We demonstrated specific values for SmO2 and PmO2 as early indicators of internal bleeding and showed that they provided significantly earlier warning than currently used clinical parameters.

We also conducted a study on sepsis patients undergoing resuscitation with the standard clinical protocol “Early Goal Directed Therapy” (EGDT) and showed that SmO2 could be used to indicate when a patient was under-resuscitated while noninvasively determined muscle pH indicated when patients were over-resuscitated with chloride-containing solutions which cause acidosis. Specific values for these parameters were established to provide noninvasively determined goals to direct treatement of septic patients.

We have developed and demonstrated accurate methods for determining muscle oxygen and pH for sick and healthy individuals independent of their skin color and fat thickness. We have demonstrated that these methods can be used on exercising individuals and have applicability for very sick patients being treated in the emergency room. These advances position us to apply a combination of these parameters for determining oxygen consumption during EVA and assessing muscle and aerobic deconditioning during space exploration.

The NIRS noninvasive metabolic monitor is expected to have many applications for NASA. The system will have additional use on earth for military and civilian personnel treating critically ill and injured patients. It can also be used in the hospital, ambulances and helicopters. As part of a Smart Medical System, advanced medical assessment and monitoring may become available to physicians in remote and rural areas, who may not have access to specialist expertise.

We developed a novel method to correct transdermal spectra for interference from overlying layers of skin and fat. This method employs a specially designed sensor and novel mathematics. This methodology was published and is patent pending. We then found that we also needed to correct for variation in muscle optical properties between subjects. We developed a new mathematical method for this and demonstrated it in a phantom study, and on pH measurements for exercising subjects. A paper on this technique has recently been published in the journal Applied Spectroscopy. A patent application has been filed.

We developed a new method for determining muscle oxygen saturation (SmO2) A paper has on has been published in the journal Optics Express and a patent application filed.

We independently validated the pH measurement by showing that anaerobic threshold calculated from hydrogen ion concentration (derived from pH) was highly correlated with the lactate threshold during a VO2max test. This paper is accepted for the Journal of Applied Physiology. A patent is pending on this technology.

2) Leg System

A system for use on the leg during cycling was completed and delivered to NASA-JSC Exercise Physiology Lab. The leg system was tested on 10 subjects doing an VO2max cycling test. A special sensor holder was designed to hold the sensor against the skin and completely eliminated motion artifacts from spectra during cycling, walking and running.

The system has been used to collect data during several treadmill tests. The first test was in the exercise physiology lab (N=10) and 2 sensors were simultaneously used, one on the calf and one on the thigh to look at the differences in SmO2 between the 2 muscles and how they contribute to whole body VO2. We also collected data from the thigh and calf during the shirtsleeve portion of the EVA Program’s Suit Test 1 and Suit Test 2 protocols in the POGO simulator. This allowed us to collect data during walking and running under simulated partial gravity conditions. Data analysis is on-going and will continue as part of the renewed project.

3) Trauma Care

We completed enough subjects to demonstrate that in patients with severe sepsis SmO2 was highly correlated with blood lactate. This very important finding provides a noninvasive method for assessing severity of illness and adequacy of resuscitation. SmO2 is an indicator of microvascular perfusion and low SmO2 indicates very poor blood flow, inadequate oxygen delivery, and consequently production of lactate. We determined on a preliminary basis that SmO2 of 40% corresponds to lactate of 4 mmol/l and suggest that a treatment goal might be to resuscitate to assure that SmO2 is significantly greater than 40%.

With the US Army we demonstrated that SmO2 and PmO2 were very early indicators of internal bleeding, highly correlated with stroke volume reduction and much earlier than changes in HR, BP and pulse oximetry.

Aviat Space Environ Med 2008 Mar;79(3):L294. , Mar-2008

Med Sci Sports Exerc. 2006;38:S248. , May-2006

American College of Sports Medicine 54th Annual Meeting , May 30 - June 2, 2007. Program and abstracts. , May-2007

During year 3 of this project we continued to validate and improve the robustness of our calibration equations for noninvasively measuring muscle pH and PO2. We also developed a new methodology for determining muscle oxygen saturation. We completed the development of a system for measuring through thicker fat layers on the thigh and delivered the system to NASA-JSC. As part of this system we developed a mechanical fixture for stabilizing the sensor on the leg and showed that measurements were immune to motion artifacts. A pilot study of 10 subjects was completed at JSC demonstrating performance of our system during a VO2max cycle ergometry test. Using data from our system we demonstrated the capability to accurately determine anaerobic threshold noninvasively from our pH measurements and demonstrated the feasibility of noninvasively determining oxygen uptake using only NIRS-determined parameters. The other intended application of the system will be in a Smart Medical System for the assessment and treatment of critically ill and injured crew. We have begun a study in the UMass Medical School Emergency Department of patients with severe sepsis. We demonstrated that our muscle oxygen saturation measurement was highly correlated with blood lactate, indicating the severity of microvascular impairment and the response of the patient to therapy. We also demonstrated good agreement of our noninvasive pH measurement with pH values determined from venous blood. We have initiated discussions with 3 potential commercial partners who have expressed an interest in working with us to develop miniature hardware which will be more suitable for spaceflight. We have also begun discussions with personnel at NASA Glenn Research Center about assistance in developing hardware which is designed to be flight certified. The NIRS noninvasive metabolic monitor is expected to have many applications for NASA. The system will have additional use on earth for military and civilian personnel treating critically ill and injured patients. It can also be used in the hospital, ambulances and helicopters. As part of a Smart Medical System, advanced medical assessment and monitoring may become available to physicians in remote and rural areas, who may not have access to specialist expertise.

1) Calibration Equations. We developed a novel method to correct transdermal spectra for interference from overlying layers of skin and fat. This method employs a specially designed sensor and novel mathematics. This methodology was published and is patent pending. We then found that we also needed to correct for variation in muscle optical properties between subjects. We developed a new mathematical method for this and demonstrated it in a phantom study, and on pH measurements for exercising subjects. A paper has on this technique has recently been published in the journal Applied Spectroscopy. A patent application has been filed. We developed a new method for determining muscle oxygen saturation (SmO2) and another new method for calculating muscle PO2 from SmO2 and the hemoglobin dissociation curve. A paper has been submitted and a patent application filed. We independently validated the PO2 measurement by showing that it was linearly related to stroke volume and inversely related to total peripheral resistance in an LBNP study simulating hemorrhagic shock. A paper on these results has been submitted. We independently validated the pH measurement by showing that anaerobic threshold calculated from hydrogen ion concentration (derived from pH) was highly correlated with the lactate threshold during a VO2max test. This paper is currently being written.

2) Leg System. A system for use on the leg during cycling was completed and delivered to NASA-JSC Exercise Physiology Lab. The leg system with a 30 mm sensor was tested on 10 subjects doing an VO2max cycling test and 2 subjects in a walk/run treadmill test. The 30 mm sensor could be used because these fit subjects had thin layers of fat over their thigh muscle. A special sensor holder was designed to hold the sensor against the skin and completely eliminated motion artifacts from spectra during cycling, walking and running.

3) Trauma Application. We completed enough subjects to demonstrate that in patients with severe sepsis SmO2 was highly correlated with blood lactate. This very important finding provides a noninvasive method for assessing severity of illness and adequacy of resuscitation. SmO2 is an indicator of microvascular perfusion and low SmO2 indicates very poor blood flow, inadequate oxygen delivery, and consequently production of lactate. We determined on a preliminary basis that SmO2 of 40% corresponds to lactate of 4 mmol/l and suggest that a treatment goal might be to resuscitate to assure that SmO2 is significantly greater than 40%. One of our prototype systems has been supplied to a collaborator at Beth Israel Deaconess Hospital in Boston who will collect additional data for this study data.

Crit Care Med. 2006;34:A54. , Dec-2006

Crit Care Med. 2006;34:A55. , Dec-2006

During year 2 of this project we improved the robustness of our calibration equations by developing methods that simultaneously corrected for spectral interferences from skin pigmentation and fat. We also developed new methodology that allowed us to correct for subject-to-subject differences in muscle optical properties. These innovations considerably increased our ability to accurately measure muscle pH, oxygen and hematocrit across all subjects.

We completed development of hardware and software to measure muscle metabolic parameters on the arm (thinner fat layer) and the leg (thicker fat layer). An arm system was delivered to the exercise physiology group at Johnson Space Center (JSC) and was used in a pilot study to develop a pre-EVA handgrip fitness test. The leg system development was completed and is currently being validated against invasive measurements at the University of Massachusetts Medical School, in a knee extension protocol. The system will be delivered to JSC in late spring.

The other intended application of the system will be in a Smart Medical System for the assessment and treatment of trauma or injury which results in reduced blood pressure, blood flow or cardiac output. We have begun a study in the UMass Medical School Emergency Department to apply the sensor to subjects at risk for developing shock. We have been able to collect data from 3 subjects who were in shock. These initial results indicate that we were able to detect changes in peripheral perfusion that accompany shock, and that the noninvasive sensor was also able to detect improvements in muscle oxygenation that paralleled standard invasive clinical measurements of improvement (blood lactate). Recruitment into this protocol is much lower than anticipated and we are beginning to explore the possibility of adding another site to the study to improve subject accrual.

During frequent trips to Johnson Space Center we have begun conversations with biomedical engineers and medical operations personnel to begin to understand the requirements to proceed to a flight study. Early conversations have indicated a likely path, though it is not clear that funds will be available to “flybadize” the hardware in the near future. We are hoping to test the device as part of the Exercise Countermeasure Program’s planned bed rest study, to validate its application in assessing loss of muscle strength in a space analog system.

The NIRS noninvasive metabolic monitor is expected to have many applications for NASA. The system will have additional use on earth for military and civilian personnel treating mass casualties. It can also be used in the hospital, ambulances and helicopters. As part of a Smart Medical System, advanced medical assessment and monitoring may become available to physicians in remote and rural areas, who may not have access to specialist expertise.

1) Calibration Equations

We developed a novel method to correct transdermal spectra for interference from overlying layers of skin and fat. This method employs a specially designed sensor and novel mathematics. We demonstrated this method on phantoms and a paper was published in Optics Letters. We also demonstrated this for the measurement of hematocrit on human subjects.

We found that we also needed to correct for variation in muscle optical properties between subjects. We developed a new mathematical method for this and demonstrated it in a phantom study, and on pH measurement of human subjects. A paper has been written for Applied Spectroscopy on this technique. An invention disclosure will be prepared shortly.

We also investigated methods to calculate the absolute concentration of capillary hemoglobin and deoxygenated hemoglobin from NIRS. This novel method also allows us to track changes in tissue water concentration and blood volume over time.

2) Leg System

We had to significantly redesign the optical system to get sufficient light from leg muscle because the thick fat layer on the leg tends to reflect most of the light away. The leg system increases the lamp power from 8W to 50W. This required a total redesign of the power management system. The leg system also required a complete redesign of the fiber optic cable. The source detector spacing was increased from 30 mm to 40 mm and the diameter of the fiber bundle in lamp cable also increased. Initial testing during knee extension dynamometry was completed using the Paratrend invasive sensors as references, however many of the sensor broke during the study, and many of the spectra were collected without proper reference measurements, making them unsuitable for calibration equation development. The leg calibration/validation study must be repeated using blood samples as a reference.

3) Trauma Application

The Emergency Department study began earlier than planned. We did complete a control study and collected data on 3 patients in shock. Subject recruitment has been slow and difficult. We recently obtained IRB permission to do the study without consent if the family is not available. This should help recruitment, but we will most likely have to involve an additional site to collect sufficient data.

Med Sci Sports Exerc. 2006 May;38(5) Suppl:S248-S249. , May-2006

Crit Care Med. 2005 December;33 (12-Suppl):A126. , Dec-2005

Proc SPIE. 2005 Nov;6007:60070, M1-6007M9. http://dx.doi.org/10.1117/12.630683 , Nov-2005

Proc SPIE. 2005 Mar;5702:104-12. http://dx.doi.org/10.1117/12.585256 , Mar-2005

Proc SPIE. 2005 Nov;6007:60070, N1-60070N8. http://dx.doi.org/10.1117/12.630646 , Nov-2005

Proc SPIE. 2005 Nov;6007:60070, O1-60070O8. http://dx/doi.org/10.1117/12.630701 , Nov-2005

During year one we developed calibration equations to relate near infrared spectra collected transdermally from the forearm of humans subjects to muscle pH, muscle PO2 and blood hematocrit. We showed that these calibration equations had the required accuracy, compared to an invasive sensor, for the intended clinical applications. We made these measurements on the forearm muscle whose strength is most important for successful completion of an extravehicular activity (EVA). We demonstrated that muscle pH is a sensitive measure of work intensity, more sensitive than blood measurements of pH or lactate. With Dr. Hagan, Manager of JSC’s Exercise Physiology Lab, we are exploring the use of the noninvasive monitor to assess astronaut fitness and readiness for EVA in space.

Also this year we developed a set of tools to help us optimize sensor and system design to be able to accurately measure the metabolic parameters on the thigh muscle, which is covered with a thick layer of fat. One of these tools is a mathematical model which describes the photon path through skin and fat to the muscle. This model was validated with a prototype sensor which had flexible source-detector spacing and a new type of tissue phantom based on a stable NIR dye which mimics deoxyhemoglobin absorbance. Sensor specifications were determined and fat correction was demonstrated on the phantom materials. Initial human studies showed that the technique is effective in removing the spectral influence of fat.

System modifications are underway to improve the monitor for adequate light detection from muscle beneath thick fat layers. In parallel, the human protocol has been redesigned to validate sensors using a leg extension model. This change was approved by NASA-JSC and will allow us to use NASA equipment for the study and validate our results against other studies in the literature. During the next year of the project we will complete validation of the sensor on the thigh, and deliver this system to NASA-JSC for evaluation during cycle ergometry in their laboratory. We conduct a weekly telecon with Dr. Hagan and have been visiting once/month. Once we have systems on-site at JSC (spring 2005) a graduate student will spend more time at JSC to support integration of the system into the exercise physiology laboratory. Additionally, we will ramp up our interaction with JSC engineering to begin to address equipment issues related to flight.

The other application of the system will be in a Smart Medical System for the assessment and treatment of trauma or injury which results in reduced blood pressure, blood flow or cardiac output. Values for muscle pH and PO2 that can be used to assess severity of injury and guide treatment will be determined from a study of patients who enter the emergency room in shock. This study is expected to begin at UMass in early fall, 2005.

The NIRS noninvasive metabolic monitor is expected to have many applications for NASA. The system will have additional use on earth for military and civilian personnel treating mass casualties. It can also be used in the hospital, ambulances and helicopters. As part of a Smart Medical System, advanced medical assessment and monitoring may become available to physicians in remote and rural areas, who may not have access to specialist expertise. The monitor is expandable to measure new chemistries by just altering the calibration equations stored in computer memory.

The phantoms and Monte Carlo model helped us identify key sensor design issues to assure optical depth penetration through fat on the forearm and the thigh. A flexible test probe was designed that allowed us to investigate a number of features such as angle and source-detector spacing. A new fiber optic probe, different from the one previously used on the palm, was implemented for exercise studies on the arm.

Previously we had used cardiac surgery patients as a subject group to vary pH, PO2 and hematocrit. However, we found that we could not consistently achieve a wide range of pH and PO2 to derive robust calibration equations. We instituted a repetitive handgrip exercise protocol and began monitoring on the forearm. This protocol allows us to compare our measurements with blood and intracellular measurements obtained by other investigators. We were also able to demonstrate the NIRS technique on moving muscle, for the first time. We completed 34 subjects, which has allowed us to develop calibration equations to calculate muscle pH and PO2 over a wide range. We are currently validating these equations on an independent group of subjects.

During the course of this study we were also able to investigate the relationship between muscle pH, PO2, PCO2 and blood values. We determined that the muscle values are more sensitive than the blood values to increases in exercise intensity and have the potential to be used as a noninvasive measurement to assess metabolic status of the muscle and the effectiveness of exercise countermeasures.

Also in this year we designed and built a prototype of a unique fiber optic probe to measure thigh muscle through thick layers of fat. Using novel mathematical methods, which we developed, we were able to demonstrate improvement in the accuracy of hemoglobin measurement on phantoms. In preliminary experiments we were able to show that the sensor was effective in correcting the spectra of anatomical features with different fat thicknesses (forearm, upper arm, calf and thigh) so they all look the same, irrespective of fat thickness.

Finally, we developed a human testing protocol for evaluating exercise on the thigh using both static and dynamic knee extension exercise. NASA-JSC will loan us their exercise equipment which is has been used to assess muscle strength before and after spaceflight.