INTRODUCTION

Top Abstract Introduction Materials & Methods Results Discussion References

AT REST, MOST OF THE INSPIRED air passes into the lungs through the nose. Several nonspecific stimuli in the inhaled air may activate mediators from nasal mucosal cells, nerves, and blood vessels, causing inflammatory reactions. The pathophysiological changes in the mucosa during an acute inflammatory reaction consist essentially of vascular alterations (17). These vascular reactions include reduced tone of the capacitance vessels, e.g., venous sinusoids, and resistance vessels, as well as enhanced transvascular fluid flux with rapidly increasing interstitial fluid volume (IFV) and exudation of fluid to the mucosal surface (18). The fluid required to form tissue edema and luminal transudate must be derived from plasma and transported across the microvascular barrier. This transcapillary fluid transport is the product of the capillary filtration coefficient and the hydrostatic and colloid osmotic pressures acting across the capillary wall. These parameters are interrelated in the "Starling equation"

(1)

(2)

where J_v is transcapillary fluid transport; CFC is the capillary filtration coefficient; c and if denote capillary and interstitial fluid, respectively;  is the capillary reflection coefficient for proteins; P and COP are the hydrostatic and colloid osmotic pressures, respectively; and P is the net filtration pressure across the capillary wall. Equation 2 shows that increased fluid flux across the capillary can be obtained by increasing the capillary filtration coefficient or net filtration pressure across the capillary wall. It is commonly accepted that increased microvascular permeability and capillary pressure are the major pathophysiological vascular events during the development of an inflammatory edema (11). Furthermore, interstitial fluid pressure (P_if) is normally considered to counteract edema formation because increased IFV will raise P_if, in turn acting as a counterpressure against further capillary filtration (1). Contrary to this accepted role for P_if, it has been observed that P_if initially decreases concomitantly with edema formation during several acute inflammatory reactions (10, 14, 21). The greater subatmospheric P_if under these circumstances is regarded as an important driving force for the initial and rapid edema formation, concomitant with increased vascular permeability. The mechanism responsible for this negative pressure seems to involve structural components of the connective tissue (23), and it has been suggested that the tissue becomes an "active" participant in transcapillary fluid exchange during the immediate part of the inflammatory reaction, thereby actively "sucking" fluid out of the vasculature (23).

The present study was performed as a first attempt to investigate the effect of acute inflammation on P_if, intravascular pressure in the venous sinusoids (P_s), and blood flow [laser-Doppler flow (LDF)] in rat nasal mucosa. We have earlier developed an experimental model for rat nasal mucosa, which allows simultaneous measurements of several vascular parameters (12). We have previously measured a nasal mucosa P_if to be ~2 mmHg (12). To our knowledge, no measurements of P_if have been carried out in the upper airways during inflammatory conditions. It has been hypothesized that in the lower airways, plasma exudation produces an increased P_if (4) and that this pressure might separate the tissue cells, allowing exuded plasma to flow between the cells and into the airway lumen. An inflammatory reaction was attained by using dextran as an inflammatory mediator because it induced an anaphylactoid reaction in rats. Dextran anaphylaxis is associated with an increased negativity of P_if in the trachea (9) and skin (21), most likely associated with degranulation of mast cells (8). Histamine was subsequently used because it is the major component of the mast cell granules and has earlier been associated with increased negativity of P_if in skin (21). The edema formation will raise P_if and correspondingly attenuate a decrease in P_if. Measurements were therefore also performed after circulatory arrest, when there would be practically no transcapillary fluid flux in response to the inflammatory agent. The pressures were measured by micropuncture technique (27), and local blood flow was measured by LDF.

	MATERIALS AND METHODS

Top Abstract Introduction Materials & Methods Results Discussion References

The experiments were performed in female Sprague-Dawley rats, weighing 200-250 g and anesthetized by intraperitoneal injections of pentobarbital sodium (Mebumal, 50 mg/ml body wt). The right femoral vein and artery were cannulated for supplemental administration of anesthetics and for blood pressure measurements, respectively. The arterial pressure was continuously recorded with a pressure transducer (model 1280, Hewlett-Packard Medical Instruments Group, Andover, MA) connected to a multichannel recorder (Gold RS3400, Ballainviers, France). The animals were breathing spontaneously and were kept in a supine position on a servo-controlled heating pad during the experiments. To eliminate movements, the head of the animal was fixed to the operating table by a stereotaxic frame with ear bars. The superior thyroid artery, a branch of the external carotid artery, was cannulated in a retrograde direction for local intra-arterial administration of drugs. Circulatory arrest was induced as part of the experimental protocol by intravenous injection of saturated potassium chloride under pentobarbital sodium anesthesia. Dextran and histamine might enhance capillary filtration, raising IFV. Thus P_if will become more positive because of fluid accumulation in the tissue and as a consequence of tissue compliance (1). Circulatory arrest was therefore induced 1 min after dextran infusion to allow distribution of dextran to the tissues, while edema formation was kept at a minimum. Otherwise, the animals were killed with saturated potassium chloride at the end of the experiments. In all experiments vascular parameters were measured simultaneously after drug challenge. The value of each studied parameter was calculated against the baseline value 1 min before the provocation.

The procedures described in this report were carried out with the approval of and in the accordance with the recommendations laid down by the Norwegian State Commission for Laboratory Animals.

Exposure of Nasal Mucosa

Measurements

P_s and P_if measurements by micropuncture technique. Pressure measurements were made by sharpened glass pipettes (tip diameter 2-4 µm) that were filled with 0.5 M NaCl colored with Evans blue and were connected to a servo-controlled counterpressure device first described by Wiederhielm et al. (27). Under guidance of a stereomicroscope, P_s (diameter 15-300 µm) and P_if could be measured after direct puncture by using a micromanipulator (Leitz, Germany). The position of the pipette tip in the vessel was verified by injection of 0.5 M NaCl solution colored with Evans blue. When the pipette tip was placed intravascularly, the dye disappeared within seconds after cessation of injection, whereas the dye persisted in the tissue for ~5-10 min after interstitial deposition. The micropipette could be inserted into the vessels and interstitium without visible distortion of the tissue and did not impede blood flow, as judged by visual inspection of red blood cell velocity.

Local blood flow measurements. The red blood cell flux in the nasal mucosa was estimated by a laser flow probe connected to a Periflux PF2 (12-kHz) laser-Doppler flowmeter (Perimed, Stockholm, Sweden). The probe tip was covered by saline and positioned close to the micropipette, 0.5 mm above and perpendicular to the exposed nasal mucosa surface. The monochromatic light (632.8 nm) from the He-Ne-laser has been shown to penetrate skin to a depth of ~1 mm (19). The laser-Doppler signal (i.e., LDF) was not influenced by local application of saline or by administration of saline into the nasal cavity in volumes equal to volumes in the experimental protocol. Output values from the laser-Doppler flowmeter might differ among animals, probably because of varying laser beam penetrations and slight differences in probe position. However, provided the probe position in individual experiments is unchanged, the output values in control measurements remained fairly stable, even in experiments of several hours' duration. Flow changes obtained after drug applications were therefore calculated as percentages of the baseline values (100%) obtained during control conditions.

Drugs

Experimental Protocol

Series I: Dextran. Dextran 40 (0.5 ml) was injected slowly (30 s) into the superior thyroid artery.

Series II: Histamine. Twenty-five microliters of histamine solution (1 mg/ml) were deposited in the nasal cavity.

Statistical Evaluation

	RESULTS

Top Abstract Introduction Materials & Methods Results Discussion References

Series I: Dextran as an Inflammatory Mediator

Intact circulation. During control conditions, P_s was 9.9 ± 1.3 mmHg. After dextran infusion, P_s increased significantly to 10.9 ± 1.4 mmHg (P < 0.01) within 10 min and reached a maximal value of 11.5 ± 1.7 mmHg (P < 0.01) 40 min after dextran infusion (Table 1). Mean arterial pressure (MAP) decreased significantly in control (P < 0.05) from 114 ± 20 to 109 ± 25 mmHg 11-20 min after dextran infusion. The decrease in MAP, from ~20 min after drug challenge, averaged 13-15 mmHg and lasted throughout the experimental period. LDF increased by 36% (P < 0.01) compared with control values (Table 1). The vasodilatatory effect was seen 10 min after dextran infusion and lasted 40 min (P < 0.05), and thereafter it began to fade. P_if in control was 2.2 ± 0.4 mmHg and increased significantly to 3.1 ± 0.5 mmHg (P < 0.01) at 30 min after dextran infusion (Fig. 1). After 60 min, P_if was 3.8 ± 0.21 mmHg. In all rats with intact circulation, an edema was apparent in the nose area of the rat; this was visualized by loss of skin markings and swelling. Some of the rats also developed respiratory distress during the experiments. Recordings of P_if in rat nasal mucosa in the same experimental setup have earlier been shown to be stable for several hours in untreated animals (12).

View larger version (15K): [in this window] [in a new window]   Fig. 1.   Effect of close intra-arterial infusion of dextran (, n = 6) and histamine intranasally (, n = 5) on interstitial fluid pressure (Pif) in rat nasal mucosa with intact circulation. Values are means ± SD. * P < 0.05 compared with control values.

Circulatory arrest. Control P_if before dextran infusion and circulatory arrest was 1.9 ± 0.6 mmHg. After dextran infusion and circulatory arrest, P_if decreased significantly to subatmospheric values, on average 0.8 ± 0.8 mmHg within 30 min after start of infusion (Fig. 2). This decrease persisted throughout the experiment. Intra-arterial saline injections had no significant effect on P_if (Fig. 2).

View larger version (22K): [in this window] [in a new window]   Fig. 2.   Effect of close intra-arterial infusion of dextran (, n = 6) and saline (, n = 5) and intranasal histamine (, n = 6) and saline (, n = 4) on Pif in rat nasal mucosa after circulatory arrest. Values are means ± SD. * P < 0.05 compared with control values.

Series II: Histamine as an Inflammatory Mediator

Intact circulation. During control conditions, P_s and MAP were 14.1 ± 2.3 and 110 ± 10 mmHg, respectively (Table 1). After intranasal application of histamine, P_s and MAP did not change significantly compared with control values. However, histamine caused an initial increase (36%) in LDF compared with control values during the first 20 min (Table 1). Thereafter, the LDF outputs remained elevated but were not significantly different from control values (Table 1). P_if also increased significantly (P < 0.05) from 1.6 ± 0.9 mmHg during control measurements to 2.9 ± 0.6 mmHg at the end of the experiment (Fig. 1).

Circulatory arrest. The control P_if before histamine challenge was 1.8 ± 0.6 mmHg. After histamine application, P_if decreased gradually to a minimum value of 3.3 ± 1.2 mmHg within 20 min (Fig. 2). The pressure then rose gradually but was still significantly lower than the control value 1 h after circulatory arrest (Fig. 2). Saline administered intranasally (n = 4) did not change P_if significantly during the experiment (Fig. 2).

	DISCUSSION

Top Abstract Introduction Materials & Methods Results Discussion References

The present study reports, for the first time, measurements of P_if in the nasal mucosa during acute inflammation in rats. When the circulation was intact, the inflammatory mediators dextran and histamine significantly increased P_if and LDF concomitantly with edema formation. Dextran also significantly increased P_s, whereas no such effect was seen with histamine. Inflammation in the nasal mucosa has been shown to increase the mucosal volume because of augmented IFV and/or blood volume (18). Because dextran infusion increases both P_s and LDF, despite MAP reduction, a considerable decrease in vascular resistance seems to take place, mainly through precapillary vasodilation. Accordingly, capillary hydrostatic pressure is increased, favoring transcapillary fluid filtration. Increased capillary fluid flux may also result from increased vascular permeability. Although the present experiments were not designed to determine this, there is substantial experimental evidence that histamine (4) or anaphylaxis induced by dextran (9) causes enhanced exudation of plasma proteins. Dextran edema is thought to be mediated by mast cell degranulation, releasing the inflammatory mediator histamine (24). Increased IFV has been measured in rat trachea and skin in dextran anaphylaxis (9, 21). Histamine acts directly on the microvascular endothelial cells via histamine receptors (7, 26), causing increased permeability and plasma extravasation (4). Taken together, it seems reasonable to suggest that the augmented IFV and edema are caused by increased capillary pressure and permeability.

Both histamine and dextran significantly increased P_if. It might therefore be speculated that the increased P_if during inflammatory reactions would favor fluid transport out of the mucosa into the nasal cavity. A similar mechanism for transepithelial bulk flow of fluid has recently been proposed by Persson et al. (20) and Gustafsson and Persson (5). They demonstrated that a rise in serosal pressure of 5 cmH₂O (~3.5 mmHg) caused transport of macromolecules across the epithelium and into the tracheal lumen in vitro. They suggested that the increased hydrostatic pressure in the mucosal tissue transiently separated the tight junctions of the tracheal epithelium to allow plasma transport along the interstitial-luminal hydrostatic pressure gradient. Thus increased subepithelial hydrostatic pressure might move plasma exudate across the respiratory epithelium through pericellular pathways (4). An analogy to this hypothesis has been proposed for the intestinal mucosa by Granger et al. (3). They described a passive fluid transport (filtration secretion) from the vasculature, through the interstitium, and into the intestinal lumen because of increased net capillary filtration pressure, in turn elevating tissue pressure, providing a low-resistance pathway for fluid transport into the intestinal lumen (3).

During control with intact circulation, P_if was consistently above atmospheric pressure, with a mean P_if ranging from 1.6 to 2.1 mmHg in the experimental groups. Inflammation induced with dextran anaphylaxis or histamine resulted in a dramatic decrease in P_if to subatmospheric values within 10 min after circulatory arrest. P_if remained subatmospheric throughout the experimental period, with mean maximal negative values being, on average, 3.3 mmHg (histamine) and 0.8 mmHg (dextran) at 20-30 min circulatory arrest. A similar response has previously been observed in inflammatory reactions in the skin (5, 21) and the trachea (9, 19). When dextran was given with an intact circulation, P_if in the rat paw skin initially decreased and then increased to positive values as edema developed (21). This decrease in P_if will act concomitantly with increased capillary hydrostatic pressure and permeability in the initial edema formation to enhance transcapillary fluid flux (23). P_if in the present experiments remained unchanged up to 60 min after circulatory arrest in the saline control groups, indicating that no or very minute amounts of fluid are removed from the interstitium during this time period. A decrease in intravascular pressure due to circulatory arrest, evaporation from the nasal cavity, or increased lymphatic flow may have dehydrated the nasal mucosa, inducing increased negativity of P_if. However, the constant P_if after circulatory arrest in the saline control group speaks against this because a continuous and gradual decrease in P_if should have been expected. By exclusion, therefore, the negative P_if measured after circulatory arrest may suggest that the interstitium may play an active role in nasal mucosal edema formation during inflammation. With intact circulation, the P_if can be active only until capillary filtration has caused fluid accumulation in the interstitium and increased P_if. With intact circulation there was no decrease in P_if after inflammatory provocation, as opposed to a decrease in P_if after circulatory arrest. A similar observation has been made in burned skin, where P_if became more negative when fluid entrance into the injured tissue was prevented (15). A low compliance in the mucosa will contribute to attenuate a decrease in P_if in the groups with intact circulation because little fluid needs to be added to attenuate the increased negativity of P_if. Consequently, after circulatory arrest, when nearly no transcapillary fluid flux takes place, swelling of hyaluronal/glycosaminoglycans gel (16) may create increased negativity of P_if because of structural rearrangements, probably perturbation of the ₁-integrins, which release the tension exerted by connective tissue cells on extracellular matrix components (22). Histamine alone induced a larger decrease in P_if than did the dextran-induced anaphylaxis, suggesting that substances other than histamine released on mast cell degranulation participate in creating an increased negativity of P_if. Alternatively, it might be a dose-response phenomenon, i.e., the amount of histamine released by the mast cells is less than that administered in the experiments.

The LDF method demonstrated increased blood flow after dextran and histamine. Local blood flow in the nasal mucosa has previously been obtained by the LDF method (12, 13). The principal advantage of the LDF method is that it is noninvasive and allows a continuous recording. However, blood flow is not being obtained in absolute units, and the LDF signal is not linearly related to blood flow unless there is a fairly constant concentration of red blood cells in the tissue (24). In the present experiments vascular congestion and tissue edema during inflammation may have influenced the LDF output values. An increased tissue hematocrit because of vascular congestion will increase the LDF signal. The average tissue blood volume in the rat nasal mucosa increases from 0.13 ml/g during decongestion to 0.25 ml/g during hyperemia and congestion (12). Although the edema that develops after inflammatory provocations may to some extent attenuate this increase in tissue blood volume, it is unlikely to be as large as the increase in vascular volume, thus validating at least the trend in the present measurements. The measurement of unchanged P_s during histamine administration is not readily explained because histamine has a vasodilating effect on the nasal mucosa in experimental animals (2, 8). The unchanged P_s might be explained by a nearly equal decrease in pre- and postsinusoidal resistance. Alternatively, too-low doses of histamine were used for topical application. However, because higher doses (unpublished observations) gave a significant decrease in systemic arterial blood pressure, this seems implausible. The lack of a histamine effect on P_s may also be explained by the fact that the histamine group in control starts at a higher P_s compared with the dextran group (Table 1), which might have blocked or obscured a possible response.

In short, the present experiments have shown that inflammation increases the nasal mucosal P_if at maintained circulation, whereas a decrease in P_if was measured after circulatory arrest. The increased P_if will be part of the net driving pressure for the promptly appearing exudation across the airway mucosa after inflammatory provocations. The decreased P_if measured after circulatory arrest suggests that structural components in the mucosa may be part of the initial development of the edema in the nasal mucosa.