AJP - Heart Fuel your research with LabChart
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Am J Physiol Heart Circ Physiol 280: H2533-H2544, 2001;
0363-6135/01 $5.00
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (74)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Braun, R. D.
Right arrow Articles by Dewhirst, M. W.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Braun, R. D.
Right arrow Articles by Dewhirst, M. W.
Vol. 280, Issue 6, H2533-H2544, June 2001

Comparison of tumor and normal tissue oxygen tension measurements using OxyLite or microelectrodes in rodents

Rod D. Braun, Jennifer L. Lanzen, Stacey A. Snyder, and Mark W. Dewhirst

Department of Radiation Oncology, Duke University Medical Center, Durham, North Carolina 27710


    ABSTRACT
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES

In this study we compare oxygen tension (PO2) histograms measured with O2 microelectrodes and a new optical PO2 measurement device, the OxyLite, in normal tissues (mouse spleen and thymus) and in tumors (R3230Ac in rats) (n = 5-6). The transient response to glucose infusion or 100% O2 breathing (hyperoxia) was also measured in tumors. PO2 histograms of spleen and thymus with the two devices were not different. The OxyLite tumor PO2 histogram, however, was left-shifted compared with the microelectrode (median PO2 1.0 vs. 4.0 mmHg, P = 0.016). Both probes responded to acute hyperglycemia with a mean increase of 3-6 mmHg, but the microelectrode change was not significant. The OxyLite consistently recorded large PO2 increases (~28 mmHg) with hyperoxia, whereas the microelectrode response was variable. The OxyLite averages PO2 over an area that contains interstitial and vascular components, whereas the microelectrode measures a more local PO2. This study demonstrates the importance of considering the features of the measurement device when studying tissues with heterogeneous PO2 distributions (e.g., tumors).

polarographic microelectrode; luminescence fiber-optic sensor; tumor hypoxia


    INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES

THE PRESENCE OF HYPOXIA IN tumors has long been known to adversely affect the sensitivity of tumors to radiation therapy (35). Hypoxia has recently been shown to increase mutation frequency (29), to exert a selective pressure for survival of those cells with lower apoptotic potential (12), to alter gene expression of genes involved in cell cycle regulation and cytokine production (6), and to impact treatment outcome and patient survival (3, 11, 15, 16, 27, 32). Thus there is a continued interest in measurement of oxygen (O2) tension (PO2) in tumors.

Although measurements of PO2 in tumors have been made using optical fluorescent techniques (14) or hypoxia markers (18), most O2 levels in solid tumors have been measured with polarographic electrodes. The electrodes have either been the needle-encased electrode used in the Eppendorf histograph system (19), gold or platinum microelectrodes (22), or recessed-glass polarographic microelectrodes (7). Recently, a new PO2 measurement system has become commercially available. This device is the OxyLite PO2 system (Oxford Optronics; Oxford, UK), which measures PO2 by using a fluorescence quenching technique. The prototype was described by Young et al. (40) in 1996 and was used to measure tumor PO2 and subsequently by Collingridge et al. (4). Briefly, the system uses a ruthenium chloride fluorescent compound immobilized in a polymer at the tip of a fiber-optic probe, which is ~220 µm in diameter. Blue light is emitted from diodes within the unit, and the light is propagated down the fiber to the tip, where it excites the ruthenium chloride. The lifetime of the resultant fluorescence is inversely proportional to the amount of O2 at the tip.

Although comparisons of tumor PO2 measurements obtained by the OxyLite and Eppendorf systems have been made (4), no detailed analysis of the differences between measurements obtained with small glass O2 microelectrodes and the OxyLite system has been made. Because microelectrodes and the OxyLite are currently being used in laboratories to measure tumor PO2, it is important to know how measurements made with the two probes are similar and how they are different.

In this study, we compared measurements of PO2 made with traditional recessed-tip O2 microelectrodes and with the OxyLite PO2 system. PO2 histograms were measured in rat tumors, mouse spleen, and mouse thymus. To fully compare the two techniques, rat tumor PO2 was also measured continuously at a single location after glucose infusion and 100% O2 breathing.


    METHODS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES

OxyLite PO2 System

The OxyLite PO2 system (Oxford Optronics) measures PO2 by determining the O2-dependent fluorescent lifetime of ruthenium chloride (40). The ruthenium chloride is immobilized at the tip of a 220-µm-diameter fiber-optic probe. Because each probe is calibrated at the manufacturer before shipment, the calibration for each probe was scanned into the computer by using a barcode wand. The probe was then ready for measurement of tissue PO2. The PO2 signal from the OxyLite probe was a 5-s average value and was recorded to disk with the use of a data-acquisition system (MacLab, ADInstruments; Castle Hill, Australia). For the PO2 response experiments, the PO2 values were averaged >10 s and graphed at 10-s intervals.

After the animal was euthanized at the end of the experiment, an in vivo zero value was obtained for each probe. If the PO2 dropped to a reasonable nonzero value, i.e., 1 or 2 mmHg, it was assumed that there was a slight calibration error and that value was used as the true zero for that probe. All PO2 measurements made in that experiment were corrected to account for the offset in the zero value.

O2 microelectrodes. Recessed-tip O2 microelectrodes were produced by using a previously published technique (24). Recess lengths were relatively short, on the order of 30 µm. Several electrodes were used for more than one experiment. The electrodes used to record PO2 histograms in mice and rats had tip diameters of 8.0 ± 2.2 µm (mean ± SD; n = 15 electrodes). The microelectrodes used in the glucose experiments had tip diameters of 8.0 ± 3.1 µm (n = 11). The O2 breathing experiments involved microelectrodes with tip diameters of 9.6 ± 3.1 µm (n = 10).

The microelectrodes were polarized at -0.7 V using a commercial polarizing box and picoammeter unit (chemical microsensor model 1201, Diamond General; Ann Arbor, MI). The signal from the microelectrode was digitized at 25 Hz and recorded using data-acquisition software (AT-CODAS; Windaq, DATAQ Instruments; Akron, OH). Electrodes were calibrated before and after each experiment in a saline-filled tonometer alternately bubbled with 0%, 5%, 15%, or 21% O2 (balance N2). The saline was warmed to 37°C. An in vivo dead value was also obtained by recording the microelectrode current in the tissue after euthanasia of the rat or mouse with an overdose of pentobarbital sodium. The average sensitivity of the microelectrodes used in the histogram studies was 1.74 ± 0.74 mmHg/pA (means ± SD; n = 15 electrodes). The average sensitivities of the microelectrodes used for the glucose and O2 breathing studies were 1.04 ± 0.85 (n = 11) and 1.00 ± 0.27 mmHg/pA (n = 10), respectively.

PO2 histograms in mouse spleen and thymus. Twenty-one female DBA/2 mice (Charles River Laboratories; Raleigh, NC) were used in this portion of the study. At the time of experiment, the mice were ~6-8 wk and weighed 16.7 ± 1.7 g. The mouse was anesthetized with an intraperitoneal injection of 80 mg/kg pentobarbital sodium. Either the spleen was exposed by an abdominal incision or the thymus was exposed via mediastinotomy. Care was taken to avoid pneumothorax. After the tissue was exposed, the tissue was kept moist by topical application of saline. For the microelectrode experiments, a small incision was then made in the left forelimb, and an Ag/AgCl reference electrode was sutured into the subcutis. Body temperature was maintained by placing the mouse on a regulated water-heated blanket (K-Module, Baxter Healthcare; Valencia, CA). For measurement of PO2 in the spleen, the mouse was placed in a lateral recumbent position. For measurements in the thymus, the mouse was positioned on its back.

After preparation of the animal and calibration of the electrode, a micromanipulator (model MO102E, Narishige; Narishige, Japan) was positioned so that a dummy probe could reach the exposed tissue surface, which was covered by a drop of saline. The actual probe or microelectrode was then placed in the micromanipulator and advanced into this saline droplet. The electrode was allowed to polarize for several minutes in the saline before being advanced into the tissue. For the OxyLite measurements, a 27-gauge needle was used to pierce the tissue and facilitate penetration of the probe. This step was not necessary for the microelectrode measurements. The OxyLite probe or microelectrode was moved into the tissue and the position on the micromanipulator was noted. The probe was then advanced in 50-µm steps for a total distance of 1,000-2,000 µm. At each location, PO2 was recorded for 10 s, and an average PO2 over the interval was calculated later. The probe was then withdrawn and reinserted at a new location. A new penetration was made, and the PO2 along the second track was recorded. This process was repeated until a total of three to four tracks had been made in the tissue. The total measurements made in each mouse ranged from 50 to 192 points.

At the end of the recording time, an overdose of pentobarbital sodium was injected intravenously. The recordings of all parameters continued for at least 5 min after death. The PO2 value after death was used in the final microelectrode calibration as a true in vivo zero as described previously (10) or was used to correct for calibration error in the OxyLite probe (see above).

Spleen PO2 was measured in 10 DBA/2 mice with a weight of 16.5 ± 1.5 g (mean ± SD). Five histograms were recorded with the OxyLite system, and five were recorded with the microelectrodes. Eleven mice were used to measure thymus PO2. The mice weighed 16.9 ± 2.0 g. Five or six experiments were performed with the use of each system.

Preparation of rats for tumor PO2 studies. Forty-six female Fischer 344 rats (Charles River Laboratories) were used in this study. All of the rats received subcutaneous implants of 1- to 2-mm3 pieces of R3230Ac rat mammary adenocarcinoma in the left hindlimb. After the tumor had reached 1 cm in diameter, the tumor-bearing rats were used in the experiments. The rat was anesthetized with an intraperitoneal injection of 50 mg/kg pentobarbital sodium. A small portion (~4-10 mm2) of the skin and tumor capsule was then removed to expose the surface of the tumor. This was kept moist by topical application of saline. For the microelectrode experiments, a small incision was then made in the left forelimb, and an Ag/AgCl reference electrode was sutured into the subcutis. Body temperature was maintained by placing the rat on a regulated water-heated blanket (K-Module, Baxter Healthcare).

After preparation of the animal and calibration of the electrode, the rat was placed on a water-heated blanket and the left leg was stabilized on a rubber pedestal with tape. Care was taken not to elevate the leg. A micromanipulator (model MO102E, Narishige) was positioned so that a dummy probe could reach the exposed tumor surface, which was covered by a drop of saline. The actual probe or microelectrode was then placed in the micromanipulator and advanced into this saline droplet. The electrode was allowed to polarize for several minutes in the saline before being advanced into the tumor. For the OxyLite measurements, a 27-gauge needle was used to pierce the tumor and facilitate penetration of the probe. This step was not necessary for the microelectrode measurements. The OxyLite probe or microelectrode was moved into the tumor and the position on the micromanipulator was noted. The signal from the microelectrode was digitized at 25 Hz and recorded with the use of data-acquisition software (AT-CODAS Windaq, DATAQ Instruments). Once the electrode was inserted into the tumor, the electrode was either advanced in discrete steps to record a PO2 histogram or the electrode was moved until a nonzero PO2 was found for the PO2 transient response experiments (see below).

At the end of the recording time, an overdose of pentobarbital sodium was injected intravenously. Recordings of in vivo zero PO2 values were recorded as described above.

PO2 histograms in rat R3230Ac tumors. Once the microelectrode or OxyLite probe was inserted into the tumor and the position had been noted, the microdrive was advanced 60 µm and then retracted 10 µm. The probe remained stationary for 5-10 s and the PO2 was recorded. The probe was repeatedly advanced in this fashion until a total penetration depth of 1,000-2,000 µm had been reached. The probe was then withdrawn outside the tumor and was repositioned for another penetration. This process was repeated until three to four tracks had been completed. This protocol resulted in histograms with a total number of points ranging from 299 to 561. Tumor PO2 histograms were measured in a total of 10 rats, weighing 169 ± 11 g (mean ± SD).

Transient PO2 responses in rat R3230Ac tumors: glucose or 100% O2 breathing. After the OxyLite probe or microelectrode had been positioned in the tumor, the probe was moved into the tissue until a clearly nonzero PO2 current was obtained, so that the PO2 had an opportunity to either increase or decrease in response to the perturbation. Because the goal of this study was to look at the response of PO2 to glucose or O2, we did not wish to leave a probe in an area that may have been chronically hypoxic with PO2 readings near zero. Once the probe had been placed, a baseline PO2 was recorded. In the case of the glucose experiments, baseline was recorded for 10 min. A glucose solution (200 mg/ml) was then infused intravenously at a rate of 0.1 ml/min. The infusion lasted from 8 to 10 min. The PO2 was recorded for 90 min after the start of the glucose infusion. The 100% O2 breathing experiments lasted 40 min. A 20-min baseline PO2 was recorded, and 100% O2 was then blown across the snout of the rat via a mask. The PO2 was recorded for 20 min while the animal breathed O2. In both experiments, the probe remained stationary during the entire recording time. After the experiments, the PO2 values were averaged over 10 s and graphed at 10-s intervals. The maximal response was defined as the highest average PO2 change over 1 min after the end of glucose infusion or the start of O2 breathing.

In the glucose experiments, 18 rats weighing 172 ± 16 g were used. Microelectrodes were used in 11 experiments, and OxyLite probes were used in the other seven. Eighteen rats weighing 168 ± 10 g were used in the hyperoxia studies. Microelectrode measurements were made in 10 rats, whereas the OxyLite system was used in the other eight experiments.

Statistics

All of the data were compared using nonparametric analysis. Differences between groups were tested with the use of the Mann-Whitney U-test. The differences in paired data were tested with the use of the Wilcoxon matched-pairs signed-ranks test. Significance was achieved if P <=  0.05.


    RESULTS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES

Tissue and Tumor PO2 Histograms

Spleen PO2. Regardless of the probe used for the PO2 measurement, qualitatively the spleen was generally well oxygenated with relatively few PO2 values near zero. The cumulative frequency plot of the histograms measured by the microelectrode showed an almost linear rise from 0 to 1.0 across a PO2 range of 4-31 mmHg (Fig. 1A). In contrast, the OxyLite recorded PO2 values mainly between 10 and 23 mmHg (Fig. 1B). Despite this difference in range, the plots show that the vast majority of the PO2 values measured by either system were between 10 and 25 mmHg. Similarly, the global histograms, including all of the measured points in all mice, revealed that most of the measured PO2 values were between 5 and 30 mmHg (Fig. 1, insets). The striking difference between the two global histograms was at the high end of the distribution. Almost 10% of the values measured by the microelectrode were >40 mmHg, whereas the OxyLite recorded <1% of its values in this range.


View larger version (25K):
[in this window]
[in a new window]
  		Fig. 1.   Distribution of PO2 in the spleen of DBA/2 mice measured with microelectrodes (n = 402 points in 5 mice) (A) or the OxyLite PO2 system (n = 466 points in 5 mice) (B). Figure shows the median cumulative frequency of the given PO2 ± interquartile range for the set of five mice. Insets: global PO2 histogram, i.e., all of the PO2 values measured in all the mice. See METHODS for details.

The characteristics of the five individual histograms determined by using the two measurement systems are compared in Table 1. None of the parameters was significantly different, although the standard deviation approached statistical significance. This could be interpreted to mean that there was a trend for the microelectrode to measure a broader range of PO2 values, which was also in evidence in the cumulative frequency plots and the global histograms (Fig. 1).

                              
View this table:
[in this window]
[in a new window]
  		Table 1.   Characteristics of PO2 histograms recorded in spleen, thymus, and tumor using oxygen microelectrodes or OxyLite PO2 system

Thymus PO2. In contrast to the distribution of PO2 in the spleen, the thymus was much more hypoxic, with many values <5 mmHg. This was true whether the microelectrode (Fig. 2A) or the OxyLite system (Fig. 2B) was used for the measurements. The large plot in each panel shows the cumulative frequencies of the individual histograms for each thymus. The cumulative frequency plots of the histograms measured by the microelectrode and the OxyLite both showed a steep rise from 0 to 0.95 across a PO2 range of 0-17 mmHg. The major qualitative difference is that over 20% of PO2 values measured by the OxyLite were <1 mmHg. This difference is shown even more dramatically when the global histograms are compared (Fig. 2; insets). Despite this difference at the lowest end of the histogram, the plots still revealed that the vast majority of the PO2 values measured by either system were between 0 and 17 mmHg.


View larger version (22K):
[in this window]
[in a new window]
  		Fig. 2.   Distribution of PO2 in the thymus of DBA/2 mice measured with microelectrodes (n = 620 points in 5 mice) (A) or the OxyLite PO2 system (n = 896 points in 6 mice) (B). Figure shows the median cumulative frequency of the given PO2 ± interquartile range for the set of 5 or 6 mice. Insets: global PO2 histogram, i.e., all of the PO2 values measured in all the mice. See METHODS for details.

The characteristics of the individual histograms determined using the two measurement systems are compared in Table 1. None of the parameters were significantly different. Thus in this tissue the two PO2 measurement systems yielded similar results.

Tumor PO2. We measured tumor PO2 with the microelectrode in five rats and with the OxyLite system in another five rats. The discrepancy between the two measurement systems was greatest in this tissue, and the difference in the two techniques is demonstrated best by looking at the cumulative frequency plots (Fig. 3). The microelectrode distribution showed a steady increase in cumulative frequency of PO2 measurements from 15.7% (median) <1 mmHg up to 80.5% at 5 mmHg and below (Fig. 3A). The OxyLite distribution started at a cumulative frequency of 53.6% <1 mmHg and then steadily rose to 84.3% at 5 mmHg (Fig. 3B). The microelectrode distinguished a more heterogeneous pattern in PO2 than could be measured by the OxyLite system. With the use of the microelectrodes, <20% of the values were <1 mmHg (Fig. 3A). By using the OxyLite probe, over 50% of the measurements were <1 mmHg in most of the tumors (Fig. 3B). If all of the measurements from the five mice were pooled into one grand histogram, only ~18% of the measurements were in this "near anoxic" range when tumor PO2 was measured with the microelectrodes (Fig. 3A, inset). The grand histogram for the OxyLite measurements showed that >50% of all measurements were in this range (Fig. 3B, inset).


View larger version (23K):
[in this window]
[in a new window]
  		Fig. 3.   Distribution of PO2 in a subcutaneously implanted R3230Ac tumor in the hindlimb of Fischer 344 rats measured with microelectrodes (n = 2,268 points in 5 rats) (A) or the OxyLite PO2 system (n = 1,628 points in 5 rats) (B). Figure shows the median cumulative frequency of the given PO2 ± interquartile range for the set of five rats. Insets: global PO2 histogram, i.e., all of the PO2 values measured in all the rats. See METHODS for details.

The characteristics of the five individual histograms determined by using the two measurement systems are compared in Table 1. Two parameters were significantly different between the two PO2 measurement devices. The median PO2 measured by the microelectrodes was significantly higher than that measured by the OxyLite probes (P = 0.016). The percentage of values <2.5 mmHg was significantly higher for the OxyLite system (P = 0.032). None of the other parameters were significantly different, indicating that the major difference between the two histograms occurred at this very low end of the distribution.

Transient Tumor PO2 Responses

Tumor PO2 response to glucose infusion. The response of tumor PO2 to an intravenous infusion of 1 g/kg glucose is shown in Fig. 4. The PO2 appeared to increase after glucose infusion using both measurement techniques. The microelectrodes detected a mean PO2 increase of ~3 mmHg after the glucose infusion (Fig. 4A), whereas the OxyLite measured a mean PO2 increase of ~6 mmHg (Fig. 4B). Whereas these responses appear similar, none of the changes in PO2 measured by the microelectrodes were statistically significant (P > 0.05, Wilcoxon ranked-sums test), although 8 of the 11 tumors showed some increase in PO2 after glucose infusion. On the other hand, the PO2 measured by the OxyLite was significantly different from baseline PO2 from 8 min (during the infusion) until 41 min (30 min after the end of the infusion) (P <=  0.05).


View larger version (24K):
[in this window]
[in a new window]
  		Fig. 4.   Mean ± SD change of rat tumor PO2 after an intravenous infusion of 1 g/kg glucose at time 0. PO2 was measured at a single point in the tumor with microelectrodes (n = 11 rats) (A) or the OxyLite PO2 system (n = 7 rats) (B). Tumor was a subcutaneously implanted R3230Ac tumor in the hindlimb of a Fischer 344 rat. Glucose (1 g/kg) was infused at time 0, and the infusion lasted 9-10 min. All changes are relative to the average PO2 1 min before the start of the glucose infusion (-1 to 0 min). For the microelectrode recording (A), none of the PO2 changes were significant (P > 0.05, Wilcoxon signed-ranks test). For the OxyLite recording (B), all changes from 8.0 to 41.0 min were significantly different from 0 (P <=  0.05).

Whereas Fig. 4 shows the mean responses and the standard deviations, it is important to note that the variability among the responses is somewhat hidden by this presentation of the data. When the maximal PO2 change is plotted as a function of baseline PO2, it is clear that the response measured by the OxyLite was much more consistent than that measured by the microelectrodes (Fig. 5). When the microelectrode was used to measure tumor PO2, 8 of the 11 measurements showed an increase in PO2 after glucose infusion (Fig. 5A). The other three actually showed a decrease of 2 mmHg or more. The mean maximum change measured with the microelectrode was 4.2 ± 7.2 mmHg (mean ± SD, n = 11). The OxyLite system consistently measured an increase of at least 6 mmHg (Fig. 5B). The mean maximum change measured with the OxyLite probe was 9.0 ± 3.2 mmHg (mean ± SD, n = 7).


View larger version (14K):
[in this window]
[in a new window]
  		Fig. 5.   Maximal change of rat R3230Ac tumor PO2 after an intravenous infusion of 1 g/kg glucose as a function of baseline PO2. Baseline PO2 is the average PO2 during the minute before the start of glucose infusion. Each point represents the maximal change in an individual rat. PO2 was measured at a single point in the tumor with microelectrodes (n = 11 rats) (A) or the OxyLite PO2 system (n = 7 rats) (B). Maximal response is defined as the highest average PO2 change over 1 min for the 30 min after the end of glucose infusion (10-40 min).

Tumor PO2 response to 100% O2 breathing. The response of tumor PO2 to 100% O2 breathing was markedly different when measured with the two techniques. When PO2 was measured with the microelectrode, PO2 did not consistently increase after 100% O2 breathing and the mean magnitude of the change was only ~10 mmHg (Fig. 6A). In addition, the increase in PO2 was not immediate, because the PO2 change from baseline PO2 was only significant after 4.3 min. On the other hand, when tumor PO2 was measured with the OxyLite system, tumor PO2 consistently increased immediately after the initiation of O2 breathing and remained elevated for the entire 20 min of gas exposure (Fig. 6B). The mean magnitude of the increase was over 25 mmHg.


View larger version (20K):
[in this window]
[in a new window]
  		Fig. 6.   Mean ± SD change of rat tumor PO2 during 100% O2 breathing. PO2 was measured at a single point in the tumor with microelectrodes (n = 10 rats) (A) or the OxyLite PO2 system (n = 8 rats) (B). Tumor was a subcutaneously implanted R3230Ac tumor in the hindlimb of a Fischer 344 rat. From -20 to 0 min, the animals were breathing air. At 0 min, 100% O2 breathing was begun and maintained for the next 20 min. All changes are relative to the average PO2 1 min before the start of O2 breathing (-1 to 0 min). For the microelectrode recording (A), the PO2 changes from 4.3 to 20 min were significant (P <=  0.05, Wilcoxon signed-ranks test). For the OxyLite recording (B), all changes after 0.7 min were significantly different from 0 (P <=  0.05).

Again, to better appreciate the difference between the responses measured by the two measurement techniques, the maximal change in PO2 for each individual experiment was plotted as a function of the baseline PO2 (Fig. 7). There was no correlation between baseline PO2 and maximal PO2 change for the microelectrode data (Fig. 7A). In 3 of the 10 cases, PO2 did not change at all or even decreased slightly. In four experiments, the maximal PO2 increase was <10 mmHg. The mean maximum change measured with the microelectrode was 17.9 ± 25.9 mmHg (mean ± SD, n = 10). Although the OxyLite data did not show a significant correlation between baseline PO2 and maximal PO2 change, there was a step-like nature to the data (Fig. 7B). One-half of the experiments showed a PO2 increase of at least 37 mmHg. In three of the other four experiments, the PO2 increase was at least 13 mmHg. In the remaining experiment, the baseline air breathing PO2 was near 0 mmHg, and the PO2 increased by ~3 mmHg. Thus if the baseline PO2 measured with the OxyLite was >0, the O2-induced change in PO2 was at least 13 mmHg. The maximum change in PO2 measured with the OxyLite probe averaged 27.8 ± 18.5 mmHg (mean ± SD, n = 8).


View larger version (14K):
[in this window]
[in a new window]
  		Fig. 7.   Maximal change of rat R3230Ac tumor PO2 during 100% O2 breathing as a function of the PO2 during air breathing. Baseline PO2 is the average PO2 during the minute before the start of 100% O2 breathing. Maximal response is defined as the highest average PO2 change over 1 min during the 20 min after the start of O2 breathing. Each point represents the maximal change in an individual rat. PO2 was measured at a single point in the tumor with microelectrodes (n = 10 rats) (A) or the OxyLite PO2 system (n = 8 rats) (B).


    DISCUSSION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES

In this study PO2 measurements using polarographic O2 microelectrodes and OxyLite optical probes were compared in three tissues and under various conditions. Measurement of PO2 histograms revealed that the results were tissue dependent. In two normal tissues, spleen and thymus, the PO2 distributions measured by the two systems were not statistically different. Differences between the histograms measured by the two techniques were found only in the very hypoxic R3230Ac tumor. In general, the distribution measured by the OxyLite was dominated by lower PO2 values and yielded a higher hypoxic fraction than the histogram determined by microelectrodes. Transient responses of tumor PO2 to two different perturbations were also examined. Although both probes showed similar small mean PO2 increases after glucose infusion, the changes measured by the microelectrode were not statistically significant. The responses of the two systems to 100% O2 breathing were drastically different. The OxyLite probe consistently showed a large increase in PO2 immediately after initiation of 100% O2 breathing. The microelectrodes recorded no significant change in PO2, sometimes showing an increase, and sometimes not changing at all. This study is the first to compare these two techniques of measuring PO2 in animal tissues in vivo. The results point to important differences between the two measurement techniques, which need to be incorporated into the interpretation of experimental O2 measurements.

PO2 Histograms in Normal Tissues

PO2 histograms in spleen. Splenic PO2 has been measured previously in rats and rabbits. Jamieson and van den Brenk (17) measured PO2 in the rat spleen using both 60- and 330-µm-diameter insulated gold wire electrodes. In an extensive study involving 140 rats, they determined a mean splenic PO2 of 17 ± 2 mmHg (means ± SE) and 23 ± 2 mmHg using the smaller and larger electrodes, respectively. In a later report, 3- to 8- µm-diameter gold microelectrodes were used to measure rabbit splenic PO2 (38). In that study, 1,054 measurements were made in 11 rabbits, yielding a grand PO2 histogram with values ranging from 20 to 100 mmHg and a mean PO2 of 63.5 mmHg. In this study, we determined a mean splenic PO2 of near 20 mmHg using either microelectrodes or the OxyLite (Fig. 1, Table 1). This value is in agreement with the earlier study in rat. A subset of these data are reported in another manuscript, and the possible significance of the O2 levels in the spleen are presented there (C. C. Caldwell et al., unpublished observations).

PO2 histograms in thymus. There have been no other measurements of PO2 in the thymus to our knowledge. Regardless of which measurement system was used, the thymus was shown to be a tissue with surprisingly low O2 levels. The mean PO2 was around 10 mmHg with a median near 8 mmHg. Whereas the mean thymic PO2 is lower than might be expected, other organs also have similarly low PO2 distributions. The mean PO2 in the vascularized half of the cat retina (near the vitreous humor) is near 13 mmHg during light adaptation (23). In some studies (25, 37), the myocardium has also been shown to have mean PO2 values as low as 5-10 mmHg. The brain may also have similarly low PO2 levels, although a wide range of PO2 values have been found (37). In a recent study (28), the cat primary visual cortex was shown to have a mean PO2 of 12.8 mmHg with 59.1% of the values <10 mmHg. The retina, myocardium, and brain are all highly metabolic tissues with sufficient blood supplies to maintain oxidative metabolism. It is not known whether the thymus is similar to these tissues or has a modest metabolism and a more limited blood supply. The former may be the case, because the mouse thymus has a dense vascular network with looped capillaries (20). As with the spleen, portions of the thymus PO2 data are reported elsewhere, and the possible significance of these low PO2 values is discussed in detail in that manuscript (C. C. Caldwell et al., unpublished observations).

Comparison of PO2 histograms in normal tissue: OxyLite vs. microelectrode. The PO2 histograms measured in normal tissues with the microelectrodes and the OxyLite probes were not statistically different (Table 1). The only parameter that suggested any difference in the distributions was the standard deviation of the histograms measured in the spleen (P = 0.056). The standard deviation of the microelectrode measurements tended to be twice as large as that measured with the OxyLite. This is consistent with the general appearance of the histograms of splenic PO2 (Fig. 1), where the range of measurements made by the microelectrodes is larger than that measured by the OxyLite. Most of the microelectrode measurements fall between 4 and 32 mmHg, with 9% of all measurements >40 mmHg. The OxyLite PO2 measurements lie primarily between 10 and 20-25 mmHg, and <1% of the measurements were >40 mmHg. This slight difference in the histograms can be explained by the fact that the OxyLite averages PO2 over a larger area than the microelectrode. The measuring tip of the OxyLite is 220 µm, whereas microelectrodes have a tip diameter near 10 µm. The effect of averaging on a relatively normal distribution would be a narrowing of the histogram, and this is consistent with the pattern seen in spleen.

PO2 Histograms in Tumors

Features of PO2 histograms in R3230Ac tumors. There have been many previous measurements of PO2 distributions in both experimental and human tumors. In general, tumors have much lower mean and median PO2 values than normal tissues, and the distributions are severely left shifted, i.e., skewed to the right, with a high percentage of the PO2 values near 0 mmHg (19, 39). The hypoxic fractions (fraction of PO2 values <5 mmHg) can vary widely, depending on such factors as tumor cell line (33) and implantation site (19).

Oxygenation of the R3230Ac tumor has been studied previously by using the Eppendorf PO2 histograph, and it has been found to be very hypoxic. The reported hypoxic fractions (PO2 <5 mmHg) for the R3230Ac range from 49 to 82% (2, 31). The hypoxic fractions measured by the OxyLite (83.4 ± 13.9%) and the microelectrodes (69.6 ± 21.4%) in the present study fall approximately within this range. The median PO2 of the R3230Ac has been reported to be 1-8 mmHg (2) and 3.6 ± 0.3 mmHg (mean ± SE) (31). These values are in agreement with those determined by the OxyLite (1.0 ± 0.7 mmHg) and the O2 microelectrodes (4.0 ± 3.5 mmHg) in the present study.

Comparison of tumor PO2 histograms: OxyLite vs. microelectrode. Tumor PO2 distributions measured by the OxyLite system and by microelectrodes were similar in that they both revealed many very low PO2 values. More importantly, however, several statistically significant differences between the distributions were found. As shown in Table 1, the median PO2 determined by microelectrodes was significantly higher than that measured by the OxyLite system. The fraction of PO2 values <2.5 mmHg was also significantly lower in the histogram measured by the microelectrodes. Thus the OxyLite measured many more PO2 values near 0 mmHg.

The first explanation for the difference in low PO2 values might be that the larger OxyLite probe had some physical effect on the tissue, and the extremely low PO2 values are artifactual. The low values could be caused by pressure on the tissue or vascular damage, both of which could reduce tumor blood flow and PO2. Although these explanations are possible, there are two pieces of evidence against this interpretation. First, the effect is only seen at PO2 values near zero. The rest of the histograms measured by the two measurement systems were similar. If the OxyLite probe decreased perfusion, one would expect to see midrange and higher PO2 values affected as well. As shown in the grand histograms (Fig. 3), the OxyLite measured as many tumor PO2 values >10 mmHg as the microelectrodes did. Second, there were no differences in the histograms for spleen and thymus. If the OxyLite probe caused damage in the tumor, it would also be expected to damage the spleen and thymus. There were no significant differences between the descriptive parameters of the histograms measured by the two different techniques in these normal tissues (Table 1). This second argument may not be valid for the pressure effect of the probe, because this effect would have been more significant in the less compliant tumor than in normal tissues. The issue of the insertion of the probes to increase pressure at the tip was recognized at the start of this study. In an attempt to minimize this effect, both probes were advanced 60 µm and then withdrawn 10 µm. Nevertheless, pressure still could have been high around the tip, and this may have contributed to some of the very low tumor PO2 values measured with the OxyLite probe.

An alternative explanation for the difference in the low end of the histograms involves the sensing volume (measurement volume) of each probe. Although the concept of "measurement volume" can be misconstrued, it obviously plays a role in determining how PO2 is measured by an O2 sensor. Because both recessed-tip O2 microelectrodes (30) and the OxyLite sensor (40) consume minimal or no O2 and do not disturb the O2 field in front of the probe significantly, the two sensors used in this study essentially measure the PO2 at the tip of the probes. The term "measurement volume" implies that the sensor measures a distinct volume, which is often taken to be hemispherical (13). A more accurate description would be that each sensor measures a region, which is a disk with a circular surface area the size of the cathode or dye tip. If the electrode consumes O2, it will disturb the PO2 field near the tip, and this could cause a hemispherical gradient to be established in the tissue (30). This would alter the true PO2 at the tip of the electrode and change the tissue PO2. With microelectrodes, the use of even a small recess reduces this problem considerably (30). By using this interpretation of the sensing volume, the microelectrode would measure an area of ~80 µm2 or less, and the OxyLite probe would measure an average PO2 in an area of ~38,000 µm2. This is a ratio of ~475:1.

To appreciate the effect of this difference in measurement area on the PO2 distribution, a simple conceptual model can be used (Fig. 8). In this model, it is assumed that the microelectrode measures individual points in the tissue and the OxyLite measures an average of 100 points. Thus the ratio here is only 100:1, but it will demonstrate the effect of PO2 averaging. In this example, a hypothetical area of a tumor is presented (Fig. 8A). The tumor section includes a central hypoxic area with small severely hypoxic regions interspersed in the field. The hypothetical distribution of individual PO2 values measured by a microelectrode is shown in Fig. 8B. This field was chosen to yield a distribution similar to that measured in the in vivo study presented earlier (Fig. 3A). When a 100-point average of the PO2 is calculated, a much different distribution results (Fig. 8C). It is characterized by a shift of the lowest PO2 values to the far left or a severe skew to the right. The percentage of PO2 values <1.0 mmHg increases from 22.1% to 44.5%, and the hypoxic fraction (% PO2 values <= 2.5 mmHg) increases from 57.3% to 84.4%. This is the same trend seen in the actual tumor data (Fig. 3, Table 1). Thus the higher hypoxic fraction measured by the OxyLite probe is not necessarily attributable to artifact, but may also be a function of PO2 averaging.


View larger version (50K):
[in this window]
[in a new window]
  		Fig. 8.   Conceptual model demonstrating effects of averaging on the PO2 distributions measured by an O2 microelectrode and an OxyLite PO2 probe. A: hypothetical PO2 field for a hypoxic tissue with a very heterogeneous PO2 distribution (e.g., tumor). Field is made up of a 25 × 25 grid of PO2 values. Grayscale shows the ranges of PO2 values in the field, with white pixels having PO2 >=  14 mmHg. B: if the microelectrode is assumed to measure at each single point in the field, then this distribution is obtained. Mean PO2 ± SD for the histogram is 3.4 ± 3.9 mmHg, with a median of 2.0 mmHg; 57.3% of the values are <2.5 mmHg. C: if the OxyLite is assumed to average the 100 surrounding PO2 values, then this distribution is obtained. Mean PO2 ± SD for the histogram is 1.5 ± 1.0 mmHg, with a median of 1.2 mmHg; 84.4% of the values are <2.5 mmHg. D: second hypothetical PO2 field for a tissue with a less random PO2 distribution. E: distribution measured by the microelectrode yields a mean PO2 ± SD of 7.4 ± 4.1 mmHg, with a median of 7.0 mmHg; 7.8% of the values are <2.5 mmHg. F: OxyLite histogram is narrower and has a mean PO2 ± SD of 5.8 ± 2.1 mmHg, with a median of 5.5 mmHg. None of the values are <2.5 mmHg.

This finding is dependent on the magnitudes of the individual PO2 values and their position in the tissue grid. This can be demonstrated by creating another PO2 field with slightly higher PO2 and a more normally distributed PO2 distribution (Fig. 8D). The hypothetical microelectrode distribution is broad and almost Gaussian in appearance (Fig. 8E). Although the PO2 values are lower, the shape of the distribution is similar to that measured in the spleen (Fig. 1A). The hypothetical OxyLite distribution is narrower and slightly left-shifted or skewed to the right (Fig. 8F). Again, this is reminiscent of what was seen in the spleen, where the OxyLite distribution (Fig. 1B) was slightly skewed to the right, compared with the microelectrode histogram (Fig. 1A).

Although this conceptual model cannot perfectly duplicate the in vivo measurements, it does clearly show that the measurement volumes of the probes can play a key role in the characteristics of the PO2 histograms measured in the tissue, particularly in a heterogeneously hypoxic tissue like tumors.

Tumor PO2 Responses to Glucose or 100% O2 Breathing

Tumor PO2 response to glucose infusion: OxyLite versus microelectrode. In tissues capable of carrying out glycolysis in the presence of O2, an overabundance of glucose stimulates these tissues to shift metabolism away from oxidative metabolism toward glycolysis (5). The shift results in a decrease in O2 consumption, which would lead to an increase in tissue PO2 if the O2 supply remains constant. This phenomenon is known as the Crabtree effect (5) and has been known to occur in tumors for more than 80 years. Thus it was hypothesized that an increase in tumor PO2 might occur after glucose infusion. A more detailed description of the effects of hyperglycemia on microelectrode PO2 is presented elsewhere (S. A. Snyder et al., unpublished observations).

When using the OxyLite probe, glucose infusion led to a statistically significant increase in tumor PO2 of ~6 mmHg. This glucose-induced increase was not artifactual, because the probe did not respond to glucose in in vitro tests (data not shown). When measured with microelectrodes, the mean PO2 increase was near 3 mmHg, but the change was not statistically significant. The difference in the results obtained with the two probes can again most likely be explained by a difference in the measurement area of the probes. The microelectrode measures PO2 in a very local area of the tumor. In 8 of the 11 microelectrode experiments, an increase in PO2 was measured at some point after glucose infusion (Fig. 5A). Half of the time the magnitude of the increase was small (<5 mmHg). If we assume that O2 consumption decreases after glucose, then we would expect the microelectrode PO2 to increase. However, if the drop in consumption were accompanied by a local decrease in blood flow (i.e., red blood cell flux), tumor PO2 might decrease (21). Hyperglycemia has been shown to increase red blood cell rigidity, increase blood viscosity, and decrease tumor blood flow (36). Although the 1 g/kg dose of glucose used in this study resulted in no change in tumor blood flow as measured by laser-Doppler flowmetry (S. A. Snyder et al., unpublished observations), there may well have been subtle changes at the microregional level. If the overall effect of this dose of glucose decreases O2 consumption and does not change tumor blood flow, then one would expect a PO2 increase if PO2 is averaged over a large area. This is the case with the OxyLite probe, which samples an area including interstitium and blood vessels. It should be remembered that vascular PO2 in the tumor would be expected to increase after glucose infusion because less O2 would be extracted from the blood as it flows through the tumor parenchyma. Therefore, both the increase in blood and parenchymal PO2 would contribute to the PO2 increase measured by the OxyLite.

Tumor PO2 response to 100% O2 breathing: OxyLite vs. microelectrode. Because the presence of O2 is important to the efficacy of radiation therapy, many studies have been performed in an attempt to increase tumor oxygenation. Most of the techniques have involved increasing the O2 supply to the tumor by increasing the O2 content of the blood. One of the simplest methods of increasing blood O2 content is having the animal or patient breathe 100% O2, and this has been studied as a potential means of increasing tumor PO2 levels (34).

Most studies looking at O2 changes in experimental tumors in response to 100% O2 breathing have measured tumor PO2 histograms with the Eppendorf electrode system and have either shown an increase in median tumor PO2 and a decrease in hypoxic fraction (34, 26) or no change in median PO2 or hypoxic fraction (2) during hyperoxia. In one study (34), transient changes at a single point were also measured using a catheter PO2 electrode with an unspecified diameter. Tumor PO2 rose from near 10 mmHg to ~70 mmHg during 100% O2 breathing. That result is similar to the OxyLite results in the current study. The OxyLite probes measured a mean increase in PO2 of over 25 mmHg during 100% O2 breathing (Fig. 6B). In all eight experiments, PO2 increased (Fig. 7B). In contrast to this, microelectrodes measured a hyperoxia-induced increase in PO2 in only 7 of 10 experiments (Fig. 7A), and the mean increase was only ~10 mmHg. Once again, this discrepancy in PO2 response to hyperoxia can most likely be attributed to differences in the sampling characteristics of the two probes. The microelectrode measures PO2 in a small region of the tumor. If the probe is near a blood vessel, there is a chance that the local PO2 will increase in response to hyperoxia. If the microelectrode is in a region at some distance from a vessel or near a vessel far from the arteriolar input, the PO2 might not change. The increased O2 content of the arterial blood may make no difference in that region of the tumor, if all of the O2 has been lost by the time it reaches that location. This extreme longitudinal O2 gradient in tumors due to limited arteriolar input makes improving oxygenation at all locations within some tumors extremely difficult (9). This situation would be exacerbated if tumors increase O2 consumption in response to hyperoxia (8). In the case of the OxyLite probe, it measures a larger area of tumor, including both vascular and interstitial components. Therefore, at least most of the vascular portion of the PO2 field over which it averages will show an increase during 100% O2 breathing. This would result in an overall increase in PO2 measured by the OxyLite probe. The heterogeneity in tumor response measured by the microelectrode may explain why breathing of high O2 content gases has shown little benefit when used in combination with radiation therapy in this model (1).

Implications of Differences Between Microelectrode and OxyLite PO2 Measurements

The results of this study point out the importance of interpreting O2 data based upon a knowledge of what exactly is being measured by the device. This appears to be particularly true for a tissue with a very heterogeneous PO2 distribution, such as a tumor. The major difference in the two systems is the averaging area of the two probes. Because the OxyLite averages over an area several hundred times larger than the microelectrode, it tends to smooth out some of the heterogeneity in the PO2 distribution. This was particularly evident in the histograms of the spleen (Fig. 1), in which the distribution was narrowed and fewer high PO2 values were recorded by the OxyLite. Despite the averaging, however, it is important to note that there were no significant differences between the distributions obtained by the microelectrode and the OxyLite in normal tissues. The averaging created significant differences only when histograms in the tumor were measured. Because the tumor is known to be hypoxic and have a heterogeneous PO2 distribution, it presents a much different situation than found in normal tissues. The effect of averaging in this tissue was to register many more severely hypoxic PO2 values with the OxyLite than with the microelectrode (Figs. 3 and 8). It is very important to note, however, that there was no difference in the percent of PO2 values <= 5 mmHg (Table 1). Thus in a severely hypoxic tissue like tumor, the OxyLite might tend to give an overestimate of the severely hypoxic fraction.

The effect of the measurement device on interpretation of the data is even more evident in the transient responses, i.e., the response of tumor PO2 to glucose infusion and 100% O2 breathing. When PO2 was evaluated with the microelectrode, both of these experiments showed that the response of tumor PO2 to the systemic changes was heterogeneous within the tumor. In some portions of the tumor, the PO2 increased as predicted, whereas in others there was no change or even a decrease. When the PO2 response was measured with the OxyLite, all of the heterogeneity was lost. There was an increase in tumor PO2 in every case. Although this was consistent with the overall trend seen with the microelectrodes, small pockets within the tumor did not show this same result. Therefore, whereas the OxyLite is extremely useful in measuring overall changes in tumor oxygenation at a regional level, it is important to remember that more subtle changes may be occurring at a microregional level.


    ACKNOWLEDGEMENTS

The authors thank Charles C. Caldwell and Mikhail V. Sitkovsky of the Laboratory of Immunology, National Institute of Allergy and Infectious Diseases, Bethesda, MD, for assistance in the measurements of spleen and thymus PO2.


    FOOTNOTES

Address for reprint requests and other correspondence: M. W. Dewhirst, Dept. of Radiation Oncology, Box 3455, Duke Univ. Medical Center, Durham, NC 27710 (E-mail: dewhirst{at}radonc.duke.edu).

The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Received 6 September 2000; accepted in final form 5 January 2001.


    REFERENCES
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES

1.   Brizel, DM, Hage WD, Dodge RK, Munley MT, Piantadosi CA, and Dewhirst MW. Hyperbaric oxygen improves tumor radiation response significantly more than carbogen/nicotinamide. Radiat Res 147: 715-720, 1997[ISI][Medline].

2.   Brizel, DM, Lin S, Johnson JL, Brooks J, Dewhirst MW, and Piantadosi CA. The mechanisms by which hyperbaric oxygen and carbogen improve tumour oxygenation. Br J Cancer 72: 1120-1124, 1995[ISI][Medline].

3.   Brizel, DM, Scully SP, Harrelson JM, Layfield LJ, Bean JM, Prosnitz LR, and Dewhirst MW. Tumor oxygenation predicts for the likelihood of distant metastases in human soft tissue sarcoma. Cancer Res 56: 941-943, 1996[Abstract/Free Full Text].

4.   Collingridge, DR, Young WK, Vojnovic B, Wardman P, Lynch EM, Hill SA, and Chaplin DJ. Measurement of tumor oxygenation: a comparison between polarographic needle electrodes and a time-resolved luminescence-based optical sensor. Radiat Res 147: 329-334, 1997[ISI][Medline].

5.   Crabtree, HL. Observations on the carbohydrate metabolism of tumours. Biochem J 23: 536-545, 1929.

6.   Dachs, GU, and Chaplin DJ. Microenvironmental control of gene expression: implications for tumor angiogenesis, progression, and metastasis. Semin Radiat Oncol 8: 208-216, 1998[ISI][Medline].

7.   Dewhirst, MW, Braun RD, and Lanzen JL. Temporal changes in PO2 of R3230Ac tumors in Fischer-344 rats. Int J Radiat Oncol Biol Phys 42: 723-726, 1998[ISI][Medline].

8.   Dewhirst, MW, Kimura H, Rehmus SWE, Braun RD, Papahadjopoulos D, Hong K, and Secomb TW. Microvascular studies on the origins of perfusion-limited hypoxia. Br J Cancer 74: S247-S251, 1996.

9.   Dewhirst, MW, Ong ET, Braun RD, Smith B, Klitzman B, Evans SM, and Wilson D. Quantification of longitudinal tissue PO2 gradients in widow chamber tumours: impact of tumour hypoxia. Br J Cancer 79: 1717-1722, 1999[ISI][Medline].

10.   Dewhirst, MW, Ong ET, Klitzman B, Secomb TW, Vinuya RZ, Dodge R, Brizel D, and Gross JF. Perivascular oxygen tensions in a transplantable mammary tumor growing in a dorsal flap window chamber. Radiat Res 130: 171-182, 1992[ISI][Medline].

11.   Fyles, AW, Milosevic M, Wong R, Kavanagh MC, Pintilie M, Sun A, Chapman W, Levin W, Manchul L, Keane TJ, and Hill RP. Oxygenation predicts radiation response and survival in patients with cervix cancer. Radiother Oncol 48: 149-156, 1998[ISI][Medline].

12.   Graeber, TG, Osmanian C, Jacks T, Housman DE, Koch CJ, Lowe SW, and Giaccia AJ. Hypoxia-mediated selection of cells with diminished apoptotic potential in solid tumours. Nature 379: 88-91, 1996[Medline].

13.   Griffiths, JR, and Robinson SP. The OxyLite: a fibre-optic oxygen sensor. Br J Radiol 72: 627-630, 1999[ISI][Medline].

14.   Helmlinger, G, Yuan F, Dellian M, and Jain RK. Interstitial pH and pO2 gradients in solid tumors in vivo: high-resolution measurements reveal a lack of correlation. Nat Med 3: 177-182, 1997[ISI][Medline].

15.   Höckel, M, Knoop C, Schlenger K, Vorndran B, Baussmann E, Mitze M, Knapstein PG, and Vaupel P. Intratumoral pO2 predicts survival in advanced cancer of the uterine cervix. Radiother Oncol 26: 45-50, 1993[ISI][Medline].

16.   Höckel, M, Schlenger K, Aral B, Mitze M, Schaffer U, and Vaupel P. Association between tumor hypoxia and malignant progression in advanced cancer of the uterine cervix. Cancer Res 56: 4509-4515, 1996[Abstract/Free Full Text].

17.   Jamieson, D, and van den Brenk HAS Electrode size and tissue pO2 measurement in rats exposed to air or high pressure oxygen. J Appl Physiol 20: 514-518, 1965[Abstract/Free Full Text].

18.   Kennedy, AS, Raleigh JA, Perez GM, Calkins DP, Thrall DE, Novotny DB, and Varia MA. Proliferation and hypoxia in human squamous cell carcinoma of the cervix: first report of combined immunohistochemical assays. Int J Radiat Oncol Biol Phys 37: 897-905, 1997[ISI][Medline].

19.   Kallinowski, F, Zander R, Höckel M, and Vaupel P. Tumor tissue oxygenation as evaluated by computerized pO2 histography. Int J Radiat Oncol Biol Phys 19: 953-961, 1990[ISI][Medline].

20.   Kato, S, and Schoefl GI. Microvasculature of normal and involuted mouse thymus: light- and electron-microscopic study. Acta Anat (Basel) 135: 1-11, 1989[ISI][Medline].

21.   Kimura, H, Braun RD, Ong ET, Hsu R, Secomb TW, Papahadjopoulos D, Hong K, and Dewhirst MW. Fluctuations in red cell flux in tumor microvessels can lead to transient hypoxia and reoxygenation in tumor parenchyma. Cancer Res 56: 5522-5528, 1996[Abstract/Free Full Text].

22.   Kruuv, J, Inch WR, and McCredie JA. Effects of breathing gases containing oxygen and carbon dioxide at 1 and 3 atmospheres pressure on blood flow and oxygenation of tumors. Can J Physiol Pharmacol 45: 49-56, 1967[ISI][Medline].

23.   Linsenmeier, RA, and Braun RD. Oxygen distribution and consumption in the cat retina during normoxia and hypoxemia. J Gen Physiol 99: 177-197, 1992[Abstract/Free Full Text].

24.   Linsenmeier, RA, and Yancey CM. Improved fabrication of double-barreled recessed cathode oxygen microelectrodes. J Appl Physiol 63: 2554-2557, 1987[Abstract/Free Full Text].

25.   Moss, AJ. Intramyocardial oxygen tension. Cardiovasc Res 3: 314-318, 1968.

26.   Nordsmark, M, Maxwell RJ, Horsman MR, Bentzen SM, and Overgaard J. The effect of hypoxia and hyperoxia on nucleoside triphosphate/inorganic phosphate, pO2, and radiation response in an experimental tumour model. Br J Cancer 76: 1432-1439, 1997[ISI][Medline].

27.   Nordsmark, M, Overgaard M, and Overgaard J. Pre-treatment oxygenation predicts radiation response in advance squamous cell carcinoma of the head and neck. Radiother Oncol 41: 31-39, 1996[ISI][Medline].

28.   Padnick, LB, Linsenmeier RA, and Goldstick TK. Oxygenation of the cat primary visual cortex. J Appl Physiol 86: 1490-1496, 1999[Abstract/Free Full Text].

29.   Reynolds, TY, Rockwell S, and Glazer PM. Genetic instability induced by the tumor microenvironment. Cancer Res 56: 5754-5757, 1996[Abstract/Free Full Text].

30.   Schneiderman, G, and Goldstick TK. Oxygen electrode design criteria and performance characteristics: recessed cathode. J Appl Physiol 45: 145-154, 1978[Abstract/Free Full Text].

31.   Song, CW, Shakil A, Griffin RJ, and Okajima K. Improvement of tumor oxygenation status by mild temperature hyperthermia alone or in combination with carbogen. Semin Oncol 24: 626-632, 1997[ISI][Medline].

32.   Sundfør, K, Lyng H, and Rofstad EK. Tumour hypoxia and vascular density as predictors of metastasis in squamous cell carcinoma of the uterine cervix. Br J Cancer 78: 822-827, 1998[ISI][Medline].

33.   Thews, O, Kelleher DK, Lecher B, and Vaupel P. Effect of cell line and differentiation on the oxygenation status of experimental sarcomas. Adv Exp Med Biol 428: 123-128, 1997[ISI][Medline].

34.   Thews, O, Kelleher DK, and Vaupel P. Tumor oxygenation under normobaric and hyperbaric hyperoxia: impact of various inspiratory CO2 concentrations. Adv Exp Med Biol 428: 79-87, 1997[ISI][Medline].

35.   Thomlinson, RH, and Gray LH. The histological structure of some human lung cancers and the possible implications for radiotherapy. Br J Cancer 9: 539-549, 1955[ISI][Medline].

36.   Traykov, TT, and Jain RK. Effect of glucose and galactose on red blood cell membrane deformability. Int J Microcirc Clin Exp 6: 35-44, 1987[ISI][Medline].

37.   Vanderkooi, JM, Erecinska M, and Silver IA. Oxygen in mammalian tissue: methods of measurement and affinities of various reactions. Am J Physiol Cell Physiol 260: C1131-C1150, 1991[Abstract/Free Full Text].

38.   Vaupel, P, Braunbeck W, and Thews G. Respiratory gas exchange and pO