Archives of Oral Biology 47 (2002) 529536

Oxygen distribution and consumption in rat lower incisor pulp

C.Y. Yu a,b,,1 , N.M. Boyd a , S.J. Cringle b , V.A. Alder c , D.Y. Yu b,c
b

School of Dentistry, The University of Western Australia, 17 Monash Avenue, Nedlands 6009, WA, Australia Centre for Ophthalmology and Visual Science, The University of Western Australia, 2 Verdun Street, Nedlands 6009, WA, Australia c Biomedicine Division, Murdoch University, 90 South Street, Murdoch 6150, WA, Australia Accepted 1 March 2002

Abstract The aim was to determine the oxygen tension (PO2 ) and rate of oxygen consumption in the pulp. Twelve rats were anaesthetised and articially ventilated. Under an operating microscope, a recessed oxygen-sensitive microelectrode was inserted into the pulp through a small saline-covered cavity on the labial surface of the lower incisor. PO2 was measured as a function of the transverse distance from the saline medium through to the middle of the pulp. Oxygen proles were characterised by a decline of oxygen tension outside the pulp in the saline medium and a steeper gradient across the interface, before a localised oxygen consuming region corresponding to the odontoblasts. A plateau with some localised uctuations was then followed by an increase in oxygen tension in the middle of the pulp. The average oxygen tension in the plateau region was 23.2 mmHg 2.1 mmHg (n = 12). A mathematical model was used to extract oxygen consumption data from PO2 proles recorded from non-perfused pulp (created by reducing systemic blood pressure). The analysis revealed that there was a distinct oxygen consumption zone in the outer pulp, which anatomically corresponded to the odontoblast layer. The average oxygen consumption rate of the odontoblasts was 3.2 0.2 ml O2 /min per 100 g pulp tissue. The zone of high oxygen consumption was 68.7 m 6.9 m (n = 24) thick. It is concluded that pulpal oxygen distribution is heterogeneous and that the odontoblast could be a major oxygen consumer within the rat incisor pulp. 2002 Elsevier Science Ltd. All rights reserved.
Keywords: Pulp; Rat incisor; Oxygen consumption; Oxygen tension; Microelectrode

1. Introduction Oxygen is essential to maintaining the vitality of the dental pulp and meeting its metabolic demands. An adequate oxygen supply is provided by the pulps rich vasculature and relatively high blood ow (Matthews and Andrew, 1995; Trowbridge and Kim, 1998). Very little is known about its oxygen distribution and consumption. Dental pulp resembles embryonic connective tissue but at its periphery is lined with a layer of highly developed cells, the odontoblasts. Unlike other hard tissue-forming cells, whose cell bodies remain close to their processes and within the tissue matrix, the odontoblast has long cellular process
Corresponding author. Fax: +61-8-9346-7666. E-mail address: chrisyu@cyllene.uwa.edu.au (C.Y. Yu). 1 Based on a thesis submitted to the Faculty of Medicine and Dentistry, The University of Western Australia, in partial fulllment of the requirements for the Ph.D. degree.

extending from the cell body a considerable distance into dentine (Holland, 1985). Odontoblasts form and maintain dentine. Active transport of uid or ions from the pulp proper to the odontoblast process within the dentinal tubules is necessary for dentine formation. Mitochondria are abundant in odontoblast cell bodies, indicating that they have a high energy requirement. A large amount of oxygen is thought to be essential for maintaining the proper function of odontoblasts in vitro (Hasegawa, 1989); their oxygen requirements in vivo have yet to be described. In general, there are three ways of studying tissue oxygen requirements. The closed-chamber technique has been used in pulp respiratory studies (Fisher, 1967; Kozam, 1968). It is an in vitro method requiring a relatively large tissue mass; as the pulp of most species is small, pooled samples of extirpated tissue are required (Taintor and Shalla, 1978; Biesterfeld et al., 1979). With the arterio-venous oxygen difference method, the isolation of the arterioles and venules required to measure oxygen content and blood ow at the

0003-9969/02/$ see front matter 2002 Elsevier Science Ltd. All rights reserved. PII: S 0 0 0 3 - 9 9 6 9 ( 0 2 ) 0 0 0 3 6 - 5

530

C.Y. Yu et al. / Archives of Oral Biology 47 (2002) 529536

tooth apex is very difcult because it is embedded in alveolar bone (Chien, 1985; Matthews and Andrew, 1995). The oxygen-sensitive microelectrode has been used to measure tissue oxygenation in vivo (Vanderkooi et al., 1991) and has yielded valuable data in various tissues and organs (Nair et al., 1986, 1987; Yu and Cringle, 2001). The response of the microelectrode is fast enough to monitor dynamic changes, it has high spatial resolution and it produces repeatable measurements. Indeed, at present, only microelectrode-based techniques are able to explore the tissue oxygen microenvironment as a function of depth (Yu and Cringle, 2001). Furthermore, quantitative oxygen consumption data have been extracted from oxygen tension data with appropriate mathematical models (Cringle et al., 1996a,b). We have now sought to describe the distribution of oxygen tension in vivo in the rat incisor pulp. An established three-layer mathematical model was used to extract the oxygen consumption data from oxygen proles recorded from non-perfused pulps.

was 110.4 mmHg 8.2 mmHg (n = 12) at the beginning of each experiment, which is in the normal range for rats. Blood gas values (pH/blood gas analyser 238; Ciba-Corning, Halstead, Essex, UK) were PO2 88.5 mmHg 4.5 mmHg, PCO2 32.3mmHg 1.5 mmHg and pH 7.45 0.02 (n = 12). Rectal temperature was continuously monitored and kept around 37.5 C with a thermostatic blanket (Harvard apparatus, Edenbridge, Kent, UK). The animal was placed on its back in a modied stellar stereotaxic device (model 51400; Stoelting, IL, USA) and its head was clamped to the stereotaxic frame. The mandible was immobilised and lower incisors were xed to the stereotaxic frame with steel rods and screw clamps. The lower incisors were splinted to the upper incisors with glass-ionomer cement (type II F; GC Fuji, Tokyo, Japan). 2.2. Oxygen-sensitive microelectrodes Recessed oxygen-sensitive microelectrodes were manufactured in our laboratory using techniques similar to those required for measuring tissue PO2 (Whalen et al., 1973). A glass pipette was lled with a low melting point alloy (Belmont 2451; Belmont Metals) to a distance of 3040 m from the tip and its internal surface was electroplated with gold. The nished electrodes had a tip diameter of 12 m and were bevelled at an angle of 17 to improve tissue penetration. The electrodes were placed in distilled water for 24 h before calibration and utilisation. Electrodes were biased at 0.7 V with respect to an AgAgCl reference electrode placed in the subcutaneous tissue of the animal. All oxygen currents were measured with an electrometer (model 487; Keithley Instruments, Cleveland, OH). Typical calibration currents in air-equilibrated physiological saline solutions were 810 pA. Considerable care was taken over the nal selection of microelectrodes for experimental use. Electrodes in which the oxygen current stabilises rapidly within a few seconds in the calibration solution give the most reliable and reproducible measurements (Yu et al., 1994). The stable calibration value was used to determine the absolute PO2 in the pulp. All calibrations were performed in a test chamber beside the animal table so that the electrical environment, cable positions and electrical ground connections were unaltered during calibration and experimental use; this also allowed a rapid calibration check immediately after removing the electrode from the pulp. 2.3. Pulpal oxygen tension measurement Under an operating microscope (OPMI 1-FC; Carl Zeiss, Oberkochen, Germany), a small labial access cavity was prepared on the lower incisor close to the gingiva with a small diamond mounted in a high speed handpiece. The cavity was deepened carefully under copious water spray until the pinkness of the pulp was visible under a thin layer of residual dentine. A modied 30G steel needle was then used to expose the pulp, often without visible bleeding. The

2. Materials and methods 2.1. Animal preparation and surgery Twelve adult SpragueDawley rats were used; their body weight ranged from 300 to 500 g. After delivery from the supplier they were acclimatised in holding cages for at least 2 weeks before any experimentation. They were fed standard laboratory rat pellets and had free access to water. Rats were anaesthetised by intra-peritoneal injection of 5-ethyl-5-(1 -methyl-propyl)-2-thiobarbiturate (Inactin; Byk Gulden, Constance, Germany), 100 mg/kg; and atropine sulphate, 20 g (Yu et al., 1994). Supplementary anaesthesia (10 mg Inactin) was used when necessary (usually twice per experiment). In a few experiments, the animal was also paralysed with a loading dose of 816 mg gallamine triethiodide (Flaxedil; May and Baker, Dagenham, UK), 40 mg/ml, into the jugular vein to eliminate body movements and spontaneous respiration. The systemic blood pressure and respiratory rate were monitored to ensure continuing deep anaesthesia. All procedures conformed to the Principles for Research involving Animals and Human Beings (Helsinki Declaration). The trachea was cannulated and the animal was articially ventilated with air at 91 breaths/min (Rodent respirator, model 683; Harvard, South Natick, MA) at a tidal volume based on body weight. The left jugular vein was cannulated for continuous monitoring of central venous pressure and for Flaxedil injection. The left femoral artery was cannulated for blood pressure monitoring, periodic bloodgas sampling and exsanguination. In some, but not all, the animals the right lingual artery was also cannulated for intra-arterial retrograde bolus injection of noradrenaline (norepinephrine) (Sigma) at a concentration of 104 M, to test the integrity of the pulpal microcirculation. Mean arterial blood pressure

C.Y. Yu et al. / Archives of Oral Biology 47 (2002) 529536

531

pinpoint exposure and the access cavity were immediately lled with physiological saline warmed to 37 C. The microelectrode was inserted into the pulp under the operating microscope. The initial orientation of the microelectrode was achieved by manual adjustment of the stereotaxic arcs. The penetration angle to the pulp surface was adjusted to be as close to perpendicular as possible. Forward and backward translations of the electrode were achieved by handset control of a piezoelectric microdrive (model 6000; Burleigh Instruments, Fishers, NY). Subsequent movement of the electrode during pulpal penetration and withdrawal was directly controlled by the computer and associated software. The computer control system also provided a continuous graphical display of all features of interest (oxygen tension, arterial blood pressure and electrode position). Before the start of measurements, the electrode was positioned with handset control under the operating microscope so that its tip just touched the pulp surface. The reproducibility of this visually judged contact position was better than 20 m. This position served as a reference point for the location of the electrode in the pulp. The electrode was then withdrawn from the pulp surface to just within the surface of the saline medium. As it was moved again toward the pulp, a drop in oxygen tension was routinely obtained even before the pulp surface was reached. This decrease in oxygen tension was due to a diffusion-depleted zone in the saline medium outside the pulp. Oxygen proles were measured in 10 m steps from just within the surface of the saline medium through to the middle of the pulp; measurements were repeated during the withdrawal of the microelectrode. A combination of visual judgement of the electrode position and the nature of the oxygen prole recorded determined the total depth of penetration. The small size and the bevel of the electrode tip, coupled with the high acceleration of the piezoelectric driver, allowed clean penetration of the pulp with minimal damage to the electrode or tissue. This was essential to achieve high reproducibility of successive measurements at the same location and a close correlation between penetration and withdrawal proles. 2.4. Oxygen consumption analysis A mathematical model was used to extract the oxygen consumption data from the oxygen proles measured in the non-perfused pulp. The model is based on that developed for oxygen consumption analysis in the avascular retina (Cringle et al., 1996a). Because the analysis can only be applied to avascular regions of tissue, the pulp was rendered non-perfused by means of controlled exsanguination. Under these conditions, the only oxygen source for the pulp was atmospheric oxygen that diffused through the saline medium. As shown schematically in Fig. 1, the mathematical model divided the pulp into three arbitrary layers from the surface to the middle. The rst was the interface layer, the second was the odontoblast cell layer and the third was the inner pulp

Fig. 1. Schematic representation of three-layer model used for pulpal oxygen consumption analysis. All of the oxygen supply to the pulp is via the atmospheric oxygen (open arrow). Q1 and Q3 (see Section 2.4) are set to zero. Best t to experimental data was used to extract the consumption rate and the thickness of Q2 . A schematic oxygen prole is superimposed on the gure showing PO2 as a function of the transverse distance (x) from the pulp surface. PS : oxygen tension at the pulp surface.

consisting of broblasts and connective tissue. The boundaries of the three layers were set at distances L1 , L2 and L3 from the pulp surface. The oxygen consumption rates in the respective layers were designated Q1 , Q2 and Q3 . According to Ficks law of diffusion, the oxygen consumption rate (Q) in a particular layer is proportional to the rate of change in oxygen ux with distance: Q = Dk d2 P dx 2

where P is the oxygen tension, x the distance and Dk is the product of the oxygen diffusion (D) and solubility (k) coefcients, which were assumed to be uniform across the pulp. Integrating this equation twice gave an equation for P as a function of distance (x): P (x) = Q 2 x + x + 2Dk

The constants and in each of the three layers were determined by applying a set of boundary conditions. This allowed the equation to be solved and the oxygen tension at any point to be expressed by the parameters PS (oxygen tension at the pulp surface), L1 , L2 , L3 , Q1 , Q2 , Q3 , D and k.

532

C.Y. Yu et al. / Archives of Oral Biology 47 (2002) 529536

Subsequently, we made the assumption that the odontoblast cell layer was the major oxygen consuming zone and we set the oxygen consumption rate of the rst and the third layers (Q1 and Q3 ) equal to zero. This simplied the analysis considerably. 2.5. Experimental protocol After access cavity preparation, a rest of approximately 60 min was taken to ensure that the pulpal blood ow returned to normal. In some animals, a single dose of 104 M noradrenaline (15 l) was injected intra-arterially as a retrograde bolus into the lingual artery before and 60 min after the cavity preparation and pulpal blood ow was measured with a laser Doppler owmeter (oLAB; Moor Instruments Limited, Axminster, Devon, UK) to conrm the integrity of the pulpal microcirculation. The tip of the probe (400 m diameter) rested on the distal surface of the lower incisor, which was isolated with a rubber dam. A pronounced decline in pulpal blood ow and a close match in the time course between the ow responses before and after pulpal exposure were always observed. Oxygen proles were measured during successive insertions into and withdrawals from the pulp in normal air-breathing animals. On completion of these measurements the microelectrode was withdrawn completely from the access cavity to avoid any possible damage to it. Systemic hypotension was then induced by controlled exsanguination at the left femoral artery. Typical blood volumes were between 5 and 10 ml. Typical systemic blood pressures following exsanguination were between 40 and 50 mmHg, at which level the pulp became non-perfused.

Successive measurements of oxygen proles were immediately repeated. 2.6. Statistics All average values are expressed as mean S.E.M. for the number of samples. For the oxygen consumption analysis we used the same set of equations and coefcients with an iterative least means-squared analysis (Sigmaplot; Jandel Scientic) to nd the best t of the model to the experimental data. The parameters PS , L1 , L2 and Q2 were extracted to give the best agreement for each oxygen prole recorded in the non-perfused pulp. The layer (L2 L1 ) was always completely within the outer pulp. 3. Results Preliminary experiments had shown that a small access cavity and pinpoint pulpal exposure has minimal effect on the integrity of the pulpal microcirculation. Furthermore, atmospheric oxygen has little inuence on the tissue oxygen tension in the inner region of the pulp. Therefore, the oxygen tensions measured in that inner region reect those in unoperated pulpal tissue. 3.1. Histology Two further rats were used for histological studies to characterise the pulp structure for comparison with the oxygen measurements. Their lower incisors were extracted and placed in 10% formalin in 0.1 M phosphates buffer (pH 7.4) and xed for 24 h at room temperature. They were

Fig. 2. Longitudinal section of a rat incisor pulp in the region between the incisal and middle third, for comparison with the pulpal PO2 prole. OD: odontoblast layer; IP: inner pulp layer; PD: pre-dentine. A large microvessel (arrow) is seen in the middle of the pulp. Haematoxylin and eosin. Bar = 50 m.

C.Y. Yu et al. / Archives of Oral Biology 47 (2002) 529536

533

Fig. 3. Typical oxygen proles measured as a function of transverse distance (m) from just within the surface of the saline medium to the middle of the pulp in a rat with normal blood pressure. Two sequential (circles and inverted triangles) penetrations (open symbols) and withdrawals (lled symbols) are plotted for the same pulpal location. PO2 proceeds with a decrease in the saline medium followed by a steeper decline across the interface, before a region of high oxygen consumption and then a plateau in the inner pulp, before a peak in the middle.

Fig. 4. Two sequential PO2 proles measured from the same location of a non-perfused pulp after exsanguination in one animal (open and closed squares). PO2 falls rapidly from 84 mmHg at the pulp surface to very low values across the inner pulp. The uctuations in the plateau and the peak in the middle of the pulp have disappeared. Two proles at normal blood pressure (110 mmHg) are also shown for comparison (open and closed circles).

534

C.Y. Yu et al. / Archives of Oral Biology 47 (2002) 529536

then de-mineralised in 10% formic acid and 5% formalin at 25 C, which usually took 710 days. The de-mineralised teeth were then embedded in parafn and sectioned longitudinally at 6 m intervals. The slices were dehydrated in graded ethanol and stained with haematoxylineosin for light microscopy. Fig. 2 is a photomicrograph from a longitudinal section of a rat incisor pulp, taken from the region between the incisal and the middle thirds where the measurements were usually performed. The odontoblasts were immediately subjacent to the pre-dentine at the periphery of the pulp. They were columnar and tightly packed. The odontoblast layer (which included the cell-rich zone) was approximately 80 m thick. The inner pulp was from the base of the odontoblast layer to the middle of the pulp; it was approximately 210 m thick. The total thickness of the xed pulp was approximately 290 m from the surface to the middle. Capillaries were seen among the odontoblasts. A large microvessel of approximately 50 m diameter was found in the microvasculature at the centre of the pulp. A small microvessel of approximately 12 m diameter was located in the inner pulp. 3.2. Pulpal oxygen distribution Fig. 3 shows the oxygen proles obtained from two successive penetrations and withdrawals in the same region of the lower incisor pulp in one animal with a mean arterial blood pressure of 110 mmHg. PO2 (in mmHg) is plotted as a function of transverse distance (in m) from just within the surface of the saline medium to the middle of the pulp. Key characteristics of the oxygen prole correlated well with our understanding of typical pulp structure and vascular distribution from histological studies. As shown in the typical prole, PO2 began with a decrease in the saline medium from 150 mmHg next to the saline surface to 85 mmHg at the pulp surface. The pulp surface was judged visually at track depth of 350 m. As the microelectrode penetrated the outer pulp, PO2 declined further to reach a minimum of 22 mmHg at a track depth of 430 m. A relatively at segment with some uctuations followed (from 430 to 610 m track depth) as the microelectrode travelled through the inner pulp, before the steep rise to a peak of 47 mmHg in the middle of the pulp (about 630 m track depth). The exact location of the minimum PO2 , the plateau, the extent of the uctuations and the peak varied considerably, presumably due to the individual nature of the vasculature in the pulp for each track. However, the same characteristics of pulpal oxygen distribution were evident in most animals. Good agreement between the penetration and withdrawal proles (Fig. 3) proved the high reproducibility of the microelectrode technique. Precise manipulation of the ultrane microelectrodes had caused minimal disturbance to the pulp microcirculation and tissue. Twelve normal pulpal oxygen proles were selected to attain the average value for pulpal tissue oxygen tension. All chosen proles had characteristics similar to those of

Fig. 5. (A) Original tracing of oxygen prole recorded from non-perfused pulp. PO2 plotted as a function of the transverse distance (m) from just within the surface of the saline medium to the middle of the pulp. The location of pulp surface was judged visually under the operating microscope to be 330 m track depth. (B) Truncated prole of (A) as a function of distance from the pulp surface. Fitted curve (solid line) is superimposed on the data points (lled circles).

the typical prole selected. Only the PO2 from the plateau in each prole were used for the calculation. The average pulpal tissue oxygen tension in the plateau region was 23.2 mmHg 2.1 mmHg. 3.3. Systemic hypotension On comparing the oxygen proles obtained from perfused (before exsanguination) and non-perfused (after

C.Y. Yu et al. / Archives of Oral Biology 47 (2002) 529536

535

exsanguination) pulp in the same animal (Fig. 4), the proles recorded from the non-perfused pulp had lost the uctuations within the plateau and the peak at the middle of the pulp. The oxygen tensions decreased rapidly from the pulp surface to very low tensions across the most of the inner pulp. The pulp was also visibly pale, providing further evidence of the shutting down of its blood supply. 3.4. Oxygen consumption analysis An example of the raw data recorded from the non-perfused pulp in a rat immediately after exsanguination and the tted curve for oxygen consumption analysis are shown in Fig. 5. The prole (Fig. 5A) was truncated at the point where the microelectrode just touched the pulp surface; data points were plotted as a function of transverse distance from the pulp surface to the middle. These points were then least mean-squares tted to the mathematical model. The tted curve is shown in Fig. 5B. In this animal, the regression analysis located a L2 L1 zone 79 m thick in the outer pulp, with an oxygen consumption rate (Q2 in Fig. 1) of 3.1 ml O2 /min per 100 g pulp tissue. A total of 24 oxygen proles recorded from non-perfused pulp were mathematically analysed and the best t parameters extracted. All proles that were free of obvious mechanical and electrical artefacts and that showed good agreement between penetration and withdrawal proles were included in the analysis. The t parameters Q2 and (L2 L1 ) were then averaged for the 24 samples. The result was an average oxygen consumption rate for the odontoblasts (Q2 in Fig. 1) of 3.2 0.2 ml O2 /min per 100 g pulp tissue, with a thickness of 68.7 m 6.9 m (n = 24).

4. Discussion This paper describes, to the best of our belief for the rst time, the oxygen prole of the rat incisor pulp. The distribution was characterised by a moderate decline in oxygen tension in the saline medium overlying the pulp and a steeper gradient in the interface before a localised oxygen consuming region, followed by a plateau with some uctuations before a rise in oxygen tension in the middle of the pulp. The features of the pulpal oxygen prole correlated well with the histological layers. The region of oxygen consumption lay within the odontoblast layer. The plateau occurred in the inner pulp; the local uctuations reected the presence of microvessels at the inner pulp. The peak corresponded to the central vasculature. The oxygen tension of rat incisor pulp depends on a number of variables that include systemic factors, such as respiration and cardiac functions, blood volume and haemoglobin, and local factors, such as local vascular perfusion and the rate of oxygen consumption. During the measurements of normal pulpal oxygen proles, blood gases and blood pressure were routinely checked and maintained within the

normal range. Every effort was taken to ensure that the integrity of pulp microcirculation was not compromised by the pinpoint exposure. Atmospheric oxygen, which penetrated the pulp via the access cavity, had minimal inuence on the deeper pulp; our preliminary study had shown that the oxygen tension was not inuenced by 100% oxygen blown over the saline surface. This can be also seen from the oxygen proles recorded from the non-perfused pulp when the internal oxygen source from the pulpal vasculature had been removed; the uctuations within the plateau region and the peak in the middle of the pulp disappeared and oxygen tensions became very low across the majority of the inner pulp, indicating that the pulpal circulation was the dominant oxygen source for the deep inner region. Therefore, the tissue oxygen distribution we measured reects a balance between the local vascular supply and local oxygen consumption in the pulp of a physiologically stable animal. The mean pulpal tissue oxygen tension in zone L2 L1 in the rat incisor pulp (Fig. 1) was 23.2 mmHg, which is lower than that found in rabbit incisor pulp (34.3 mmHg) (Kozam, 1967). The disagreement could be attributable to the different species, microelectrodes and measuring techniques used. The microelectrode used in Kozams study had a much bigger tip (2 mm diameter) and the access cavity was over the apical third of the rabbit incisor through the mandibular bone. Spatial analysis of oxygen proles has been used to extract oxygen consumption data from different layers of the retina (Cringle et al., 1996a). The analysis requires the tissue under investigation to be avascular, i.e. having no internal oxygen sources. Because the dental pulp is highly vascular, the circulation must be shut down for an oxygen consumption analysis. We produced an avascular pulp by inducing systemic hypotension via controlled exsanguination; this has been shown to be an effective way of drastically reducing pulpal blood ow, even to total cessation (Kim et al., 1980). There are two possible reasons for cessation: severe vasoconstriction of the pulpal vessels due to the baroreceptor-activated sympathetic reex stimulation (Tonder, 1975) and a considerable reduction in perfusion pressure due to a fall in systemic blood pressure (Sasano et al., 1989). The great majority of external atmospheric oxygen appeared to be consumed in the mitochondria-rich odontoblast and cell-rich layers. The average rate of oxygen consumption by the odontoblasts (Q2 in Fig. 1) was 3.2 0.2 ml O2 /min per 100 g pulp tissue. As the pulp is enclosed in a rigid chamber, it has to be exposed for the microelectrode to gain access. The labial access cavity preparation would inevitably cause some injury to the odontoblasts. Therefore, it is arguable that the oxygen consumption rate may not be a normal physiological value. Furthermore, there is no source of atmospheric oxygen in the normal unexposed pulp. Nonetheless, the ndings do provide an indication that odontoblasts are capable of a high oxygen consumption rate, which may be similar to that of the brain (Chien, 1985). Pulpal blood ow is relatively high and also similar to that of the brain (Kim et al., 1980). The high

536

C.Y. Yu et al. / Archives of Oral Biology 47 (2002) 529536 Hasegawa, N., 1989. Effects of various culture conditions on matrix formative functions of rat incisor odontoblasts in a pulpdentine slice culture system. Jpn. J. Oral Biol. 31, 392403. Holland, G.R., 1985. The odontoblast process: form and function. J. Dent. Res. 64, 499514 (Special issue). Kim, S., Fan, F., Chen, R.Y., Simchon, S., Schuessler, G.B., Chien, S., 1980. Symposium 3: effects of changes in systemic hemodynamic parameters on pulpal hemodynamics. J. Endod. 6, 394 399. Kozam, G., 1967. Oxygen tension of rabbit incisor pulp. J. Dent. Res. 46, 352358. Kozam, G., 1968. Fluctuations in the rate of oxygen uptake of rat incisor pulp. J. Dent. Res. 47, 392394. Matthews, B., Andrew, D., 1995. Microvascular architecture and exchange in teeth. Microcirculation 2, 305313. Matthews, B., Hughes, S.H., 1988. The ultrastructure and receptor transduction mechanisms of dentine. Prog. Brain Res. 74, 6976. Nair, P.K., Buerk, D.G., Whalen, W.J., Schubert, R.W., 1986. Two cytochrome oxygen consumption model and mechanism for carotid body chemoreception. Adv. Exp. Med. Biol. 200, 293300. Nair, P.K., Buerk, D.G., Halsey Jr., J.H., 1987. Comparisons of oxygen metabolism and tissue PO2 in cortex and hippocampus of gerbil brain. Stroke 18, 616622. Sasano, T., Kuriwada, S., Sanjo, D., 1989. Arterial blood pressure regulation of pulpal blood ow as determined by laser Doppler. J. Dent. Res. 68, 791795. Taintor, J.F., Shalla, C.L., 1978. Comparison of respiration rates in different zones of rat incisor pulp. J. Dent. 6, 6370. Tonder, K.H., 1975. The effect of variations in arterial blood pressure and baroreceptor reexes on pulpal blood ow in dogs. Arch. Oral Biol. 20, 345349. Trowbridge, H.O., Kim, S., 1998. Pulp development, structure and function. In: Cohen, S., Burns, R.C. (Eds.), Pathways of the Pulp. Mosby, St. Louis, pp. 386424. Vanderkooi, J.M., Erecinska, M., Silver, I.A., 1991. Oxygen in mammalian tissue: methods of measurement and afnities of various reactions. Am. J. Physiol. 260, C1131C1150. Whalen, W.J., Nair, P., Ganeld, R.A., 1973. Measurements of oxygen tension in tissues with a micro oxygen electrode. Microvasc. Res. 5, 254262. Yu, D.Y., Cringle, S.J., 2001. Oxygen distribution and consumption within the retina in vascularised and avascular retinas and in animal models of retinal disease. Prog. Retin. Eye Res. 20, 175208. Yu, D.Y., Cringle, S.J., Alder, V.A., Su, E.N., 1994. Intra-retinal oxygen distribution in rats as a function of systemic blood pressure. Am. J. Physiol. 267, H2498H2507.

pulpal blood ow may be due to signicant arterio-venous shunting rather than all being a nutrient blood ow. Their high metabolic rate indicates that the odontoblasts are functionally active and may play an important part in the pulp. Because rat lower incisors are continuously growing, odontoblasts in the middle region may actively engage in the formation of secondary dentine. Other possible functions, such as involvement in transmitting sensory impulses, have also been suggested (Matthews and Hughes, 1988). In conclusion, we have demonstrated the heterogeneity of pulpal oxygen distribution and that the odontoblasts may be the major oxygen consumers in the rat incisor pulp. As tissue oxygen tension is a sensitive indicator for the pulp under stress, extending the microelectrode technique to investigate alterations in the pulpal tissue oxygen environment in the early stages of inammation would be of interest.

Acknowledgements We thank Mr. Dean Darcey for his technical assistance, Mr. Syd Adam for his histological preparations, Ms. Vanessa Moscarda for her various preparatory works and Ms. Sheree Hunt for her secretarial assistance. This research was supported by a postgraduate scholarship from the National Health and Medical Research Council of Australia, and an Athelstan and Amy Saw medical research fellowship from the University of Western Australia.

References
Biesterfeld, R.C., Taintor, J.F., Marsh, C.L., 1979. The signicance of alterations of pulpal respiration: a review of literature. J. Oral Pathol. 8, 129139. Chien, S., 1985. Hemodynamics of the dental pulp. J. Dent. Res. 64, 602606 (Special issue). Cringle, S., Yu, D.Y., Alder, V., Su, E.N., Yu, P., 1996a. Modelling oxygen consumption across an avascular retina. Aust. N. Z. J. Ophthalmol. 24, 7072. Cringle, S., Yu, D.Y., Alder, V., Su, E.N., Yu, P., 1996b. Oxygen consumption in the avascular guinea pig retina. Am. J. Physiol. 271, H1162H1165. Fisher, A.K., 1967. Respiratory variations within the normal dental pulp. J. Dent. Res. 46, 424428.