This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Similar articles in Web of Science Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Web of Science (10) Citing Articles via Google Scholar Google Scholar Articles by Edelman, G. Articles by Hoffman, W. E. Search for Related Content PubMed PubMed Citation Articles by Edelman, G. Articles by Hoffman, W. E.

---

NEUROSURGICAL ANESTHESIA

Cerebral Venous and Tissue Gases and Arteriovenous Shunting in the Dog

Department of Anesthesiology, University of Illinois at Chicago, Chicago, Illinois

Address correspondence and reprint requests to William E. Hoffman, PhD, Department of Anesthesiology, M/C 515, University of Illinois at Chicago, 1740 West Taylor St., Chicago, IL 60612. Address e-mail to whoffman{at}uic.edu

	Abstract

Top Abstract Introduction Methods Results Discussion References

 
Cerebral venous blood gas values have been used to indicate brain tissue oxygenation. However, it is not clear how cerebral tissue and venous measures may vary under physiologic conditions caused by arteriovenous shunt. The purpose of this study was to measure brain tissue and local cerebral venous oxygen (PO₂) and carbon dioxide (PCO₂) partial pressure during changes in ventilation and to calculate shunt fraction. Eight dogs were anesthetized with isoflurane. After a craniotomy, a Neurotrend probe (Diametrics Inc., St. Paul, MN) that measures PO₂, PCO₂, pH, and temperature was inserted into brain tissue, and a small vein that drained the same tissue was catheterized. Arterial, cerebral venous, and brain tissue PO₂ and PCO₂ were measured during random changes in ventilation to produce five different levels of inspired oxygen (room air, 40%, 60%, 80%, 95%) at each of three different end-tidal PCO₂ (20 mm Hg, 40 mm Hg, 60 mm Hg). Arteriovenous shunt was calculated from oxygen and CO₂ content in artery, vein, and tissue, representing capillary. Tissue PCO₂ was 8 mm Hg greater than vein PCO₂ during hypocapnia and this difference increased to 20 mm Hg during hypercapnia. Vein PO₂ was 8 mm Hg higher than tissue PO₂ during hypocapnia, and this difference increased to 40 mm Hg during hypercapnia. Shunt fraction increased from 10%–20% during hypocapnia to 50%–60% during hypercapnia. These results show that brain vein and tissue PO₂ and PCO₂ differ because of arteriovenous shunting and this difference is increased as end-tidal PCO₂ increases.

Implications: We found, in dogs, that the gradient between brain venous and tissue PO₂ and PCO₂ is increased with increased arterial PCO₂. The divergence between tissue and venous gases can be described by arterial to venous shunting.

	Introduction

Top Abstract Introduction Methods Results Discussion References

 
Cerebral venous blood, particularly jugular bulb measurements, had been used to evaluate brain tissue oxygenation in normal and brain-injured patients (1,2). However, the relationship between cerebral venous and tissue measures may vary with the ventilation state of the patient because of brain arterial to venous shunting. Using radioactive microspheres, it has been estimated that shunting of blood flow in brain vessels <13 µm in diameter is 10%–20%, and this fraction increases nonsignificantly to 25% during hypercapnia (3,4). Physiologic shunting of oxygen and carbon dioxide may also occur if diffusion is inhibited or tissue metabolic demand/perfusion mismatch is present (5). The purpose of this study was to measure cerebral venous and tissue oxygen (PO₂) and carbon dioxide (PCO₂) partial pressure during physiologic changes in ventilation. Shunt fraction was calculated from oxygen and CO₂ content in artery, vein, and tissue (representing capillary) (6).

	Methods

Top Abstract Introduction Methods Results Discussion References

 
This study was approved by our institutional animal care committee, and experiments were performed at the West Side Veterans Administration Animal Research Facilities in Chicago. Eight nonpurpose-bred dogs were used in this study. The dogs were fasted overnight. On the day of the study, the dogs were anesthetized with IV thiopental 25 mg/kg, the trachea intubated, and the lungs ventilated with 2% isoflurane in an air/oxygen mixture (inspired oxygen fraction [FIO₂] = 40%). Brain temperature was maintained at 38°C during the study. Catheters were inserted into the femoral artery and vein for blood pressure recording, blood gas sampling, and fluid and drug administration. Sterile saline was infused 4 mL · kg^-1 · hr^-1 IV for fluid maintenance. A 4-cm diameter craniotomy was performed over the left hemisphere, and the dura was incised and retracted.

A Neurotrend probe (Diametrics Inc., St. Paul, MN) was calibrated on the day of the study using precision gases. The probe is 0.5 mm in diameter, and four sensors measuring pH, PCO₂, PO₂, and temperature are positioned at the end with a separation of 0.5 cm. The probe was inserted into the cortex 1-cm deep, parallel to the surface of the brain. A small vein on the surface of the brain that appeared to provide drainage in the territory of the probe was catheterized with PE50 tubing. It was assumed that homogeneous tissue responses occurred within a brain region with global changes in ventilation. These procedures were performed with little or no bleeding or injury to local tissue. When surgery was completed, the craniotomy site was overlaid with sterile gauze soaked with saline to exclude extraneous light without compressing the cortical surface. End-tidal isoflurane concentration was decreased to 1.7%, and the dogs were allowed to stabilize for 30 min.

The respiratory rate and tidal volume were adjusted to produce three target end-tidal CO₂ concentrations: 1 = 20 mm Hg, 2 = 40 mm Hg, 3 = 60 mm Hg, in random order. At each CO₂ value, FIO₂ was adjusted by changing the oxygen/air mixture to five different levels: 1 = room air, 2 = 40%, 3 = 60%, 4 = 80%, 5 = 95%, in random order. Each dog was tested at all CO₂ and FIO₂ values. The equilibration time at each ventilation variable was 10 min.

At the end of each respective equilibration period, tissue PO₂, PCO₂, pH, and temperature were recorded. Arterial and local cerebral venous blood samples were obtained to measure blood gases and pH, using an Instrumentation Laboratories 1202 Blood Gas Analyzer (Lexington, MA). At the end of the study, the dogs were killed using a euthanasia solution.

Physiologic shunt fraction was calculated in each animal from oxygen content and separately from CO₂ content in the artery, vein, and capillary. The Neurotrend probe, which provides a measure of PO₂ and PCO₂ within a tissue region of 1–2 mm, was used to estimate capillary gases in a similar manner to that reported for the lungs (6). This is based on the three-dimensional architecture of arteries, veins, and capillaries in the brain and the fact that each sensor would provide a physiologic average relating to these vessels (7). Oxygen content was calculated from PO₂, in each case using the oxygen dissociation curve for the dog with a correction for pH and temperature (8,9). Venous blood pH was used rather than tissue pH for the capillary calculation. Whole blood CO₂ content was estimated from the CO₂ dissociation curve for whole blood, with a correction for oxygen saturation (10). Shunt fraction was calculated as follows: O₂ shunt fraction = (vein-capillary)/(artery-capillary) x 100; CO₂ shunt fraction = (capillary-vein)/(capillary-artery) x 100.

Data are reported as mean ± SD. End-tidal CO₂ and FIO₂ treatments were compared using a two-way repeated measures analysis of variance. A P value < 0.01 was considered significant. A Pearson Product Moment correlation test was used to evaluate the relationship between venous and tissue measures of CO₂ and O₂.

	Results

Top Abstract Introduction Methods Results Discussion References

 
Cardiovascular and arterial blood gas values at each ventilation state are shown in Table 1. Blood pressure, heart rate, and brain temperature were statistically similar among the different treatment conditions. Using analysis of variance, there was a significant effect of end-tidal CO₂ on PaCO₂, a significant effect of FIO₂ on arterial PO₂ (PaO₂), and a significant interaction between end-tidal CO₂ and FIO₂ on PaO₂. The average blood hemoglobin concentration was 13 ± 1 /100 mL.

View larger version (23K): [in this window] [in a new window]   Figure 1. Brain-tissue venous oxygen (PO2) (top) and carbon dioxide (PCO2) (bottom) differences with increases in end-tidal CO2 (ETCO2) and arterial PO2 (PaO2). There was a significant effect of ETCO2 on both PCO2 and PO2 differences but no effect of PaO2 as determined by repeated measures of analysis of variance. The error bars indicate SD.

View larger version (22K): [in this window] [in a new window]   Figure 2. Shunt fraction, calculated by O2 content (top) and CO2 content (bottom). There was a significant effect of end-tidal CO2 on CO2 and O2 shunt fraction as indicated by repeated measures of analysis of variance. There was no significant effect of PaO2 on shunt fraction. The error bars indicate SEM.

	Discussion

Top Abstract Introduction Methods Results Discussion References

We used a tissue probe that has been validated previously for continuous intraarterial monitoring of blood gases (11). Baseline measures of brain tissue PO₂, PCO₂, and pH vary between studies (12–14), and our results suggest this is related to differences in ventilation and oxygenation. Other confounding factors include different animal species and the depth of sensor insertion in the cortical tissue (15). Menzel et al. (12) reported that, in pigs, PO₂ was higher in the sagittal sinus than tissue. When FIO₂ was increased to 100%, PO₂ in sagittal sinus and tissue increased, and the correlation between these was r = 0.96. The authors observed that there was heterogeneity in tissue PO₂ measures and that sagittal sinus gases were less variable. This is consistent with our results. The venous sample represents an average of heterogeneous tissue PO₂ and shunting of the tissue drained. Our results show a correlation for tissue O₂ and venous O₂ overall measures, r = 0.72, P < 0.01. In preliminary studies, we found that sagittal sinus PO₂ was higher than in smaller veins on the surface of the cortex. This suggests that arteriovenous shunting or extracranial contamination increases as the cerebral vein becomes larger. We would expect that the correlation between venous and tissue samples would decrease as the draining vein becomes larger and more distant from the tissue measured.

Previous investigators have evaluated arteriovenous shunt by other methods. Mariani et al. (16) reported that, in patients, shunting of 25- to 50-µm particles was <1%, whereas shunting of vascular malformations and vascular tumors ranged between 50% and 100%. Fan et al. (4) found that shunting of 9-µm microspheres in dogs was 20% during normocapnia and increased to 25% when CO₂ was added to the inspired gases. Marcus et al. (3) also reported shunting of 7- to 10-µm microspheres. However, in both studies, shunting of 15-µm microspheres was <2%. Our results show that the shunt fraction calculated from blood gases increases from hypocapnia to hypercapnia. Based on previous reports using microspheres, we assume shunting occurs in 6- to 13-µm vessels in the brain. However, the physiologic shunt fraction in our calculations is different from the shunt fraction measured with microspheres because other factors, such as the velocity of blood flow or metabolic demand/perfusion mismatch, can affect tissue gas exchange in addition to perfusion of noncapillary shunt vessels.

One of our assumptions in calculating shunting was that tissue measures of PO₂ and PCO₂ are indicative of capillary values. This is similar to assumptions made when using end-tidal gas measures to estimate capillary gases in the lung (5,6). Because of the three-dimensional matrix of capillaries in the brain, it is likely that the PO₂ and PCO₂ sensors of the probe are in close association with arterial, capillary, and venous vessels, and the measured gases represent an average of this association (7). In addition, the CO₂ gradient between brain tissue and blood is <1 mm Hg (17). However, the oxygen diffusion coefficient is 1/20 that of CO₂, and this may be a factor of error underestimating capillary PO₂ from tissue measurement. This error appears to be small because the average shunt fraction at each end-tidal CO₂ level is similar for the O₂ and CO₂ calculations. This suggests that in the normal brain, tissue PO₂ and PCO₂ measures with the Neurotrend provide an estimation of capillary gases.

In conclusion, our results show that cerebral venous O₂ is correlated to tissue O₂ and the difference between these two can be interpreted by arteriovenous shunting. Cerebral venous blood overestimates tissue oxygenation, and the difference is dependent on the ventilation state. Shunt fraction increased from 10%–20% during hypocapnia to 50%–60% during hypercapnia using both oxygen and CO₂ in the blood for the calculation. Increases in shunt fraction with hypercapnia suggest that treatments that enhance cerebral blood flow may simultaneously increase shunting. It is likely that these responses would be altered by cerebral pathology.

	Acknowledgments

 
We thank Rick Ripper, CVT, for his technical expertise and experience in completing this study.

	References

Top Abstract Introduction Methods Results Discussion References

 

1. Matta BF, Lam AM, Mayberg TS, et al. A critique of the intraoperative use of jugular venous bulb catheters during neurosurgical procedures. Anesth Analg 1994;79:745–50.[Abstract/Free Full Text]
2. Tiagarajan A, Goverdhan PD, Chari P, Somasunderam K. The effect of hyperventilation and hyperoxia on cerebral venous oxygen saturation in patients with traumatic brain injury. Analg 1998;87:850–3.[Abstract/Free Full Text]
3. Marcus ML, Heistad D, Ehrhardt JC, Abboud FM. Total and regional cerebral blood flow measurement with 7–10, 15, 25 and 50 µm microspheres. J Appl Physiol 1976;40:501–7.[Abstract/Free Full Text]
4. Fan FC, Schuessler GB, Chen RY, Chien S. Determinations of blood flow and shunting of 9 and 15 µm microspheres in regional beds. Am J Physiol 1979;237:H25–33.
5. Milhorn HT Jr, Pulley PE Jr. A theoretical study of pulmonary capillary gas exchange and venous admixture. Biophys J 1968;8:337–45.
6. Wagner PD. Ventilation perfusion relationships. Ann Rev Physiol 1980;42:235–68.[Web of Science][Medline]
7. Fennema M, Wessel JN, Faithfull NS, Erdmann W. Tissue oxygen tension in the cerebral cortex of the rabbit. Biol 1988;248:451–60.
8. Rossing RG, Cain SM. A normogram relating PO₂, pH, temperature and hemoglobin saturation in the dog. J Appl Physiol 1966;21:195–201.[Free Full Text]
9. Severinghaus JW. Oxyhemoglobin dissociation curve correction for temperature and pH variation in human blood. J Appl Physiol 1958;12:485–6.[Free Full Text]
10. Bock AV, Field H Jr, Adair GS. The oxygen and carbon dioxide dissociation curves of human blood. Biol Chem 1924;59:353–77.
11. Venkatesh B, Brock TH, Hendry SP. A multiparameter sensor for continuous intra-arterial blood gas monitoring a prospective evaluation. Crit Care Med 1994;22:588–93.[Web of Science][Medline]
12. Menzel M, Rieger A, Roth S, et al. Simultaneous continuous measurement of pO₂, pCO₂, pH and temperature in brain tissue and sagittal sinus in a porcine model. Acta Neurochir Suppl 1998;71:183–5.[Medline]
13. Hoffman WE, Charbel FT, Edelman G. Brain tissue oxygen, carbon dioxide and pH in neurosurgical patients at risk for ischemia. Anesth Analg 1996;82:582–6.[Abstract]
14. Bacher A, Kwon JY, Zornow MH. Effects of temperature on cerebral tissue oxygen tension, carbon dioxide tension and pH during transient global ischemia in rabbits. Anesthesiology 1998;88:403–9.[Web of Science][Medline]
15. Dings J, Meixensberger J, Jager A, Roosen K. Clinical experience with 118 brain tissue oxygen partial pressure catheter probes. Neurosurgery 1998;43:1082–95.[Web of Science][Medline]
16. Mariani L, Haldemann AR, Schroth G. Arteriovenous shunt measurement during endovascular therapy for cerebrospinal lesions. Am J Neuroradiol 1997;18:1679–89.[Abstract]
17. Ponten U, Siesjö BK. Gradients of CO₂ in the brain. Physiol Scand 1967;67:129–40.

This article has been cited by other articles:

				K. A. Strauss, J. Lazovic, M. Wintermark, and D. H. Morton Multimodal imaging of striatal degeneration in Amish patients with glutaryl-CoA dehydrogenase deficiency Brain, July 1, 2007; 130(7): 1905 - 1920. [Abstract] [Full Text] [PDF]

				D. J Doolette, R. N Upton, and C. Grant Perfusion-diffusion compartmental models describe cerebral helium kinetics at high and low cerebral blood flows in sheep J. Physiol., March 1, 2005; 563(2): 529 - 539. [Abstract] [Full Text] [PDF]

				A. D. McLeod, F. Igielman, C. Elwell, M. Cope, and M. Smith Measuring Cerebral Oxygenation During Normobaric Hyperoxia: A Comparison of Tissue Microprobes, Near-Infrared Spectroscopy, and Jugular Venous Oximetry in Head Injury Anesth. Analg., September 1, 2003; 97(3): 851 - 856. [Abstract] [Full Text] [PDF]

				W. E. Hoffman, G. Edelman, R. Ripper, and H. M. Koenig Sodium Nitroprusside Compared with Isoflurane-Induced Hypotension: The Effects on Brain Oxygenation and Arteriovenous Shunting Anesth. Analg., July 1, 2001; 93(1): 166 - 170. [Abstract] [Full Text] [PDF]

				K. Yoshitani, M. Kawaguchi, N. Sugiyama, M. Sugiyama, S. Inoue, T. Sakamoto, K. Kitaguchi, and H. Furuya The Association of High Jugular Bulb Venous Oxygen Saturation with Cognitive Decline After Hypothermic Cardiopulmonary Bypass Anesth. Analg., June 1, 2001; 92(6): 1370 - 1376. [Abstract] [Full Text] [PDF]

				W. E. Hoffman and G. Edelman Isoflurane Increases Brain Oxygen Reactivity in Dogs Anesth. Analg., September 1, 2000; 91(3): 637 - 641. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Similar articles in Web of Science Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Web of Science (10) Citing Articles via Google Scholar Google Scholar Articles by Edelman, G. Articles by Hoffman, W. E. Search for Related Content PubMed PubMed Citation Articles by Edelman, G. Articles by Hoffman, W. E.