email

Comparison of EPR Oximetry with Two Commonly Used Methods for Assessment of Tissue Oxygenation
during Steady State and Respiratory Challenge

Julia O’Hara¹, Huagang Hou¹, Eugene Demidenko²,
Nadeem Khan¹, Hongsheng Yu¹, Roger Springett¹, Harold M. Swartz¹

¹EPR Center for the Study of Viable Systems, Department of Diagnostic Radiology,
Dartmouth Medical School, Hanover, NH 03755, USA;
²Section of Biostatistics and Epidemiology, Dartmouth-Hitchcock Medical Center, Lebanon, NH 03756, USA

INTRODUCTION

The goal of this study is to provide comparisons of EPR oximetry with methods commonly in use or currently under development for measurement
of tissue oxygenation. In this study we compared measurements of rat brain cortex pO₂ obtained using EPR oximetry with those made using
either the Eppendorf microelectrode or the Oxylite fluorescence based system. To optimize the utility of the data obtained, we used each method
in the manner in which it is commonly used. That is, for EPR oxygen sensitive material (LiPc) was implanted and then a period of healing allowed
before making measurements. This is possible because non-invasive measures of pO₂ can be made after the implantation. To optimize the potential
for good agreement and to fully utilize the capability of EPR oximetry to make non-invasive measurements once the material is implanted,
we made measurements with EPR and Eppendorf or Oxylite in the same animal. In the initial study EPR and Eppendorf measurements
were made during normoxia, with sequential measurements. We also made simultaneous EPR and Eppendorf as well as EPR and Oxylite
measurements during normoxia and a change in FiO₂ to 95%O₂:5%CO₂.

METHODS

Animals: Fischer CDF, 200-225g, male, artificially ventilated and cannulated for blood gas sampling. All measurements were made under normal physiologic
parameters of MABP, HR, temperature, p_aO₂, p_aCO₂, and blood saturation.

EPR Oximetry: A chronic implant of oxygen sensitive material (LiPc, <100mg) was stereotactically placed on the left side of the brain of rats
at a depth of about 1.5-2 mm below the dura via a 25ga needle equipped with a plunger. Measurements were made using an L-band (1.2GHz)
spectrometer using a loop resonator placed above the skull. Care was taken to avoid artificial line broadening. In most cases,
single scans of 7s were used. Linewidths were converted to pO₂ using a calibration curve for the specific LiPc used.

Eppendorf pO₂ Histograph 6650: The Ag/AgCl reference electrode was attached to the ear of the rat. The calibrated needle electrode (0.27mm dia)
was inserted via a burr hole in the skull and moved from the dura vertically into the cortex, taking measurements at 0.5 mm intervals to a depth of 4.5 mm.
Track 1 was on the left side, adjacent to the LiPc implant. Tracks 2 and 3 were on the contralateral side. The time period for the Eppendorf
measurements were about the same as that required for the EPR measurements (total about 10 minutes), which includes setting up of the animal.

Oxylite: The Oxylite system depends on the oxygen sensitivity of a chromophore fixed to the end of a fiber designed to remain in place
and provide measurements over time. In this case the EPR and Oxylite measurements were made simultaneously on the same side of the brain.
We compared both absolute tissue pO₂ during normoxia and hyperoxia, as well as relative pO₂ during a change in FiO₂ from normoxia (26%O₂)
to elevated oxygen using carbogen (95% O₂: 5%CO₂).

Data Analysis: An ANOVA model was used to analyze the variance of Eppendorf and EPR techniques and to determine the relationship
between the methods. T-tests were used to compare means of values between Oxylite and EPR measurements. Significance level was p <0.05.

RESULTS

A comparison of absolute Eppendorf and EPR measurements demonstrated: 1) That when comparing values averaged over the groups of animals,
the two methods reported similar but statistically different values with the EPR method reporting a somewhat higher average pO₂; 2)
There was poor correlation between the values measured with the two techniques in the same animal on the same side of the brain; 3)
The Eppendorf method reported a larger range of values of point pO₂; 4) Consideration of the distribution of the heterogeneity of oxygen levels
in the brain and the volumes sampled by the two methods provide an adequate explanation for most of the observed differences; 5)
The differences in the average pO₂ measured by the two methods are consistent with other studies that suggest that the Eppendorf electrode may,
under some circumstances, report a depressed pO₂, perhaps due to local perturbations from the method.

 Results from the sequential assessment of brain pO₂ with Eppendorf and EPR in the same brain are shown in Figure 1. The density of Eppendorf
values is derived from analysis of the frequency histogram of tissue PtO₂ values, using local smoothing by the Gaussian kernel with the default window
(S plus 6.1, Insightful, Seattle, WA) (1). For example, the probability of having values of PtO₂ from 10 to 30 is the area under the density/curve,
approximately 0.02*(30-10) = 40%.

The method for use of the Oxylite allows for placement of the probe in one site and continuous monitoring of brain temperature and PtO₂.
We established a routine of taking a baseline measurement of cortex PtO₂ with EPR and then stereotactically implanting the tip of the Oxylite optode
in proximity to the LiPc. Oxylite readings were initially very low compared to later values and to baseline EPR made before optode insertion
and rose steadily after insertion of the optode, requiring 30-60 minutes for stabilization. Comparisons were made during 10 minutes intervals 1)
after stabilization of the PtO₂, 2) while the rat was breathing carbogen (followed for 15-25 minutes after changing breathing gas) and 3) during recovery,
about 30-40 minutes after switching back to normoxic conditions. Both methods showed similar absolute values and changes in brain PtO₂.
Results are shown in Figure 2 above. There was a difference in baseline (p =0.0002), but there was no difference during carbogen or during recovery period.
In 9 rats tested, all responded to the changes in FiO₂ with rapid and consistent increases in tissue PtO₂. In 2 of the animals,
the Oxylite signal became unstable and therefore no value could be obtained for PtO₂ during the carbogen period.
Therefore, these animals were excluded from analysis.

CONCLUSION

These results indicate that the three methods provide different views of oxygen in tissue, and therefore it is difficult and probably not usefu
to attempt to characterize one of these as “better”. Overall they reported similar values for average PtO₂ in the cortex of rat brain.
They differ in the volumes that are sampled in a single measurement, the number of measurements, and the occurrence of trauma associated
with the measurement. The Eppendorf histograph reveals tissue heterogeneity that cannot be resolved by EPR oximetry or the Oxylite
because of the smaller volume that is a sampled by the Eppendorf. The Eppendorf and Oxylite techniques inherently can cause trauma
at the time that the measurement is made. The EPR method provides a means to make repeated measurements without closely
 associated trauma but does not resolve small regions of heterogeneity.

REFERENCE

(1). Silverman, B. W. (1986). Density Estimation for Statistics and Data Analysis. Chapman and Hall, London.

ACKNOWLEDGEMENTS

This work was supported by a NIH (NIBIB) grant PO1 EB002180, “Measurement of pO₂ in Tissues In Vivo and In Vitro,”
and used the facilities of the EPR Center for the Study of Viable Systems supported by NIH (NIBIB) grant P41 EB002032.

back to research list
home