INTRODUCTION

Top Abstract Introduction Materials & Methods Results Discussion References

PRIMARY FUNCTIONS of the small intestine include absorption of nutrients and water and maintenance of proper fluid balance in the body. Microcirculatory networks within the intestinal villi facilitate these vital functions by regulating the exchange of proteins, principally albumin, between the blood and interstitium. Albumin maintains colloidal osmotic pressure (COP) and serves as a transport vehicle for ingested fatty acids (7). Despite these important physiological activities, the mechanisms by which mucosal capillaries regulate albumin exchange are not fully understood.

Investigative findings on the effects of acute blood stasis (10 min) on endothelial surface charge and ultrastructure may provide insight into the regulatory mechanisms for intestinal albumin exchange. First, when blood flow to the intestine is stopped for 2-10 min, cationized ferritin binds to the capillary luminal surface, whereas binding does not occur following brisk blood flow (3). This information suggests that acute blood stasis causes the capillaries to express a negative surface charge not normally present and further raises the possibility that mucosal capillaries modulate an electrostatic barrier to albumin. Second, 10 min of blood stasis reduces the amounts of native ferritin seen in the interstitium after injection into the intestinal circulation (2). Native ferritin, like albumin, is a negatively charged protein. However, acute blood stasis does not alter amounts of native ferritin in endothelial plasmalemmal vesicles. Hence, stasis-induced reduction in native ferritin passage does not appear to involve vesicular transport. Finally, acute stasis reduces the density of the fenestrae with the pore sizes conferred by fenestral diaphragms becoming more varied in those that remain (8). Taken together, the modifications in capillary surface charge and fenestrae suggest that albumin exchange will be reduced following acute blood stasis.

Mucosal capillaries may physiologically experience acute stasis, because the intestine is prone to wide fluctuations in blood flow, and reduced flow to organs such as skeletal muscle leads to temporary stasis in a greater percentage of the capillaries (6, 13). During times of fast and exercise, when intestinal flow is lower, mechanisms may operate to reduce the interstitial uptake of albumin, since its transport functions are not needed in the gut and albumin retention in the vasculature would promote water absorption.

The purpose of this investigation was to test the hypothesis that acute blood stasis reduces the interstitial uptake of albumin from intestinal capillary networks. To achieve this aim, we flushed the intestinal circulation free of blood, perfused the intestine with fluorescent albumins after blood flow and stasis treatments, and then determined fluorescence intensities within capillaries and adjacent interstitial areas by computer analysis of videotapes made by recording multiple mucosal capillary networks. After normal blood flow, a diffuse pattern of fluorescence appeared in the interstitium adjacent to the capillaries and produced low-contrast images. In comparison, after 10 min of blood stasis, the interstitial areas were dark, while the capillaries remained bright. In these high-contrast images following acute blood stasis, the measured interstitial fluorescence intensity values were lower than those obtained after blood flow. Neither the high-contrast images nor lower interstitial intensity values associated with acute blood stasis were seen after acute stasis when either saline or plasma COP-equivalent saline was substituted for blood. These results are consistent with the hypothesis that acute blood stasis reduces the interstitial uptake of albumin from intestinal capillaries.

	MATERIALS AND METHODS

Top Abstract Introduction Materials & Methods Results Discussion References

Animals

Chemicals and Solutions

Surgical Procedure

Next the intestine was prepared for observation of blood flow in the mucosa. A 2-cm segment of ileum (near the second Peyer's patch proximal to the ileocecal junction for consistency) was cauterized longitudinally, slit, spread open, flushed with HBS to clear intestinal contents from the mucosal surface, and positioned mucosal side up over the pedestal of a microscope stage to view capillary networks. This stage had been specifically designed for viewing the mucosal microcirculation, while supporting the rat and maintaining animal body temperature at 37°C. The intestinal preparation was kept moist by suffusing, at a flow rate of ~0.2 ml/min, a few drops of warm HBS containing 2% BSA and 1 µM isoproterenol.

At the start of each experiment, the mucosal preparation was observed microscopically (Zeiss Axioplan, ×32 with 0.60 numerical aperature objective) to verify normal blood flow, which was defined by the blurred appearance of briskly moving blood cells. Villi adjacent to Peyer's patches were selected for capillary observation. These particular villi appeared as flat, leaflike structures rather than pillars, and thus the contained capillary networks were easier to visualize. Previous observations of the intestinal preparations after intravascular fluorescent BSA injection showed uniform extravasation of the tracer, irrespective of whether villi were adjacent to, or remote from, Peyer's patches.

Experiments

View larger version (9K): [in this window] [in a new window]   View larger version (22K): [in this window] [in a new window]   Fig. 1.   Diagrams of experimental protocols showing general two- treatment scheme (A) and specific experimental protocols (B). COP, colloidal osmotic pressure.

Acute blood stasis. To determine whether the interstitial uptake of albumin was reduced after 10 min of blood stasis, the rats were divided into three experimental groups, each having two treatments: group 1 consisted of five rats with treatment 1 having normal flow and treatment 2 having blood stasis; group 2 had four rats and treatment 1 was blood stasis and treatment 2 was normal flow; group 3 consisted of four rats and both treatment 1 and treatment 2 had normal flow. The protocols for these groups are illustrated in Fig. 1B.

After normal flow in treatment 1, group 1 rats, blood was cleared by clamping the aorta at the base of the SMA and flushing 3 ml of HBS containing heparin sulfate through the cannula at 100 mmHg. Once the microcirculation was clear of blood, the microscope illumination was switched from bright field to epifluorescence, and then 3 ml of FITC-BSA were injected and retained by applying a second clamp to the HPV. Immediately when we detected fluorescence in the capillaries through the microscope, a video timer (For-A Company, Japan) was started, and images were recorded using a video camera. Pressure was measured through the aortic cannula after the second clamp was applied to retain the FITC-BSA.

At the end of the video-recording period after treatment 1, both clamps were released to reinstate blood flow. After the 10-min flow recovery period, treatment 2 was started by applying both clamps to produce acute blood stasis, which was verified by observing stationary blood cells in the capillaries. After the 10-min stasis treatment, the HPV clamp was removed and blood cleared as before. This time TRITC-BSA was injected. Again, images of villus capillary networks were recorded.

In group 2 rats, treatment 1 consisted of a 10-min blood stasis period, followed by 10 min of normal blood flow to the intestine as treatment 2. In group 3 rats, the intestine was not exposed to blood stasis for either treatment 1 or 2. Instead, fluorescent BSAs were injected and images recorded after two consecutive 10-min brisk flow treatments.

In all groups, the order of fluorescent BSA was switched from one experiment to the next. Furthermore, additional experiments were performed to document images on 35-mm film, which offers higher resolution than that obtained on the videotapes, for presentation purposes.

Saline stasis. Two control experiments were performed to test the effect of stasis on interstitial albumin uptake, when blood was replaced by HBS, as shown in Fig. 1B. In these experiments, fluorescent BSA was injected following an initial normal blood flow period as treatment 1. For treatment 2, blood was flushed from the intestine with 3 ml HBS, and vascular clamps were applied to retain the stationary HBS within the microvasculature. After the 10-min treatment, the venous clamp was removed and the second fluorescent BSA was injected.

Plasma COP-equivalent saline stasis. Three experiments were performed in which FITC-BSA was injected following a 10-min period of stasis with plasma COP-equivalent HBS as treatment 1 as illustrated in Fig. 1B. After recording was completed, the vascular clamps were removed to reinstate blood flow. Ten minutes later, the aortic and HPV clamps were applied to produce acute blood stasis as treatment 2. After a 10-min period of blood stasis, the clamps were removed, blood was cleared, and TRITC-BSA was injected.

Data Collection and Analyses

Video recording and photographic techniques. Recordings were made for the first minute after injection, and then for 10-s intervals every 20 s for the next 5 min. During the recordings, villi adjacent to the Peyer's patches were randomly scanned, and a number of villus networks were brought to focus. No attempt was made to record the same villi from one time point to the next or after the two treatments. The scanning technique was used to increase the number of capillaries and adjacent interstitial areas sampled to more accurately determine interstitial uptake of albumin in the mucosal segment as a whole. Furthermore, we hoped to avoid damage to the villus networks, which might increase capillary leakiness, by minimizing the illumination times.

Because the leaflike villi were rarely oriented exactly perpendicular to the line of focus, the whole microvascular network of a particular villus was not in focus simultaneously. For example, if the central region was in focus, the edge region, which contains the feeding arteriole, was not in focus. The unfocused arterioles created fuzzy fluorescence, which appeared to be associated with the interstitial tissue. Transient refocusing established that the arteriolar fluorescence was not interstitial but was due to tracer retained within the vessels.

When a 35-mm camera was used to capture images, photographs were taken approximately every 30-45 s up to 6.5 min after injection of fluorescent BSA following treatments. This time frame differed from the video-recording technique, because it took longer to focus, take pictures, and advance film in the camera. The random scanning technique was again used to select villus networks.

Data collection from videotapes. Data were collected off-line by replaying the tapes and using a frame-grabber, analog-to-digital converter and computer software (NIH Image) to measure fluorescence intensity within specific areas traced in the stopped frames. Within a given frame, capillaries in sharp focus within a particular villus were first outlined, and then the computer program was used to obtain capillary intensity (I_c).

After we obtained the I_c values in a particular frame, the areas outside the capillaries, but within the villus, were traced to obtain an interstitial intensity (I_i) value. The values of N, which accompany the mean I_i values obtained in this study, refer to the number of intensity measurements that were made in each case. Regions close to arterioles were excluded from this analysis, because out-of-focus arterioles might contribute erroneously to interstitial fluorescence, as discussed above. The I_i measurements were obtained from interstitial areas outside capillaries in four to eight villi for five time intervals (0-30, 45-60, 120-130, 180-200, and 300-310 s) after each fluorescent BSA injection.

Three preliminary experiments demonstrated that the arterioles retained fluorescence intensity throughout the experiments, regardless of treatment. I_i values measured outside the arterioles after normal flow and after 10 min of blood stasis were 40.9 ± 3.2 AIU (means ± SD, N = 36) and 41.8 ± 2.9 AIU (means ± SD, N = 36), respectively, and were not significantly different from each other. Therefore, the arterioles were excluded from the analysis.

Furthermore, early experiments demonstrated that interstitial fluorescence during particular time points after treatments in specific experiments did not vary much from villus to villus. For example, after one normal flow treatment, nine interstitial measurements that were made in separate villi between 186 and 194 s after injection of fluorescent albumin yielded an average fluorescence intensity of 61.28 ± 8.22 AIU (mean ± SD).

Data analysis. To standardize for differences in intensity between FITC-BSA and TRITC-BSA in the same experiment, the following procedure was implemented. First, mean I_c and I_i from treatment 1 were plotted versus time. Next, mean I_i values for each time point of treatment 2 were multiplied by the ratio of mean I_c values for treatment 1 to mean I_c values for treatment 2. These normalized I_i values were then analyzed.

For the analysis, all measurements obtained for each treatment in each experimental group over the 6-min time period were pooled, and then means were determined (see Table 2). The pooled I_i means were analyzed by one-way repeated-measures analysis of variance (ANOVA) using Statistical Analysis System software. When differences were detected, contrast was determined. Statistical significance was set at the 0.05 level.

To examine interstitial albumin uptake over time, all I_i values obtained were pooled under the separate time intervals 0-30, 45-60, 120-130, 180-200, and 300-310 s for treatments 1 and 2 in each experimental group. These data were presented versus time for comparison of the mean I_i values for the various time intervals after the two treatments in experimental groups (see Fig. 4). The beginning time point for each time period over which values were averaged was used to plot data. For example, after we averaged the measurements collected between 45 and 60 s, the mean was plotted at 45 s on the x-axis of the graph.

	RESULTS

Top Abstract Introduction Materials & Methods Results Discussion References

Acute Blood Stasis Experiments

View larger version (53K): [in this window] [in a new window]   Fig. 2.   Microcirculatory network of an intestinal villus viewed by epifluorescence microscopy after injection of fluorescein isothiocyanate (FITC)-bovine serum albumin (BSA) into the intestinal circulation. These networks are usually formed by single arterioles (arrows), which extend from villus base to apex, to branch into capillaries (arrowheads) that interconnect. Bar = 50 µm.

Figure 3 demonstrates the different patterns of fluorescence that were associated with normal blood flow (Fig. 3, A-D) and acute blood stasis (Fig. 3, E-H). The images shown were made during a group 1 rat experiment in which observations were captured on 35-mm film using the scanning technique to record multiple villus networks over time. Within seconds after fluorescent BSA injection following intestinal flow (treatment 1), a cloudy or diffuse fluorescence appeared outside the capillaries (Fig. 3A, arrowheads; 27 s post-FITC-BSA injection). The interstitial murkiness, which suggested albumin leakage, quickly spread throughout the villi (Fig. 3B; 1 min, 37 s post-FITC-BSA injection). Seepage of fluorescence was not localized to distinct points or to specific vessels but instead occurred along the capillary length. After 3 min, it was often difficult to focus on individual capillaries due to loss of contrast between the capillaries and the interstitium (Fig. 3C; 3 min, 32 s post-FITC-BSA injection) unless red blood cells (RBCs) were identified in the intestinal circulation. Occasionally, RBCs were seen in some capillaries (Fig. 3D, arrowheads; 6 min, 17 s post-FITC-BSA injection) perhaps due to failure of the initial HBS flush to remove all RBCs or due to the possibility that RBCs from the venous circulation made their way past the HPV clamp in a retrograde manner. The presence of stationary cells, when it occurred, was associated with higher capillary fluorescence intensity (compare Fig. 3, C and D), which is suggestive of albumin retention.

View larger version (119K): [in this window] [in a new window]   Fig. 3.   Images obtained by epifluorescence microscopy during a representative experiment on group 1 rats after an initial flow treatment followed by FITC-BSA (A-D), then a subsequent 10-min stasis treatment followed by tetramethyl rhodamine isothiocyanate (TRITC)-BSA injection, showing albumin retention within capillaries (E-H). Bright areas in E and G are due to "out of focus" arterioles (as described in text) and do not demonstrate interstitial uptake. A: at 27 s post-FITC-BSA injection, leakage of fluorescent albumin begins (arrowhead). B: by 1 min, 37 s post-FITC-BSA injection, fluorescence fills villus (V). C: 3 min post-FITC-BSA injection. D: if capillaries contain stationary red blood cells (arrowheads), capillaries retain intense fluorescence even after 6 min, 17 s post-FITC-BSA injection and adjacent interstitium is dark (compare D with C). Capillaries exposed to 10-min stasis do not leak fluorescent BSA. Interstitium is darker than seen in A-C after normal flow. E: 29 s post-TRITC-BSA injection. F: 1 min, 32 s post-TRITC-BSA injection. G: 3 min, 23 s post-TRITC-BSA injection. H: 6 min, 20 s post-TRITC-BSA injection. Bar = 100 µm.

In the same experiment, a very different pattern of fluorescence was observed after 10 min of blood stasis (treatment 2 in this group of rats): the capillaries were very bright, while surrounding areas remained dark after TRITC-BSA injection (Fig. 3, E-H), irrespective of whether RBCs were identified in capillaries. Villi that contained cell-free capillaries were dark similar to those capillaries containing RBCs. The diffuse bright areas seen in Fig. 3, E and G, were not due to interstitial staining; they illustrate the effects of out-of-focus arterioles. Changing the plane of focus revealed that the arterioles retained the TRITC-BSA within the lumen. Even 6 min after the injection of fluorescent BSA, capillaries exposed to acute blood stasis retained a high degree of fluorescence intensity, with little fluorescence appearing in the interstitial areas, and were similar in appearance to those photographed within a minute or two after BSA injection (compare Fig. 3, F and H). These high-contrast images, due to bright capillaries surrounded by dark interstitial areas after fluorescent BSA injection following acute blood stasis treatments, are consistent with the hypothesis that interstitial albumin uptake is reduced when mucosal capillaries are subjected to acute blood stasis.

To quantify the different patterns of fluorescence seen after normal flow and acute stasis, fluorescence intensities were measured within and outside of the vessels using a computer program to analyze images from the videotapes. Mean I_i data and order of the fluorescent BSA following the two treatments for individual rats in the acute blood stasis experiments are presented in Table 1. Mean I_i values for data pooled by experimental groups appear in Table 2.

Analysis of the data in Table 2 using one-way repeated-measures ANOVA detected differences among the mean I_i values obtained from the experimental groups. Using contrast, differences between treatment 1 and treatment 2 achieved statistical significance in groups 1 and 2 (P = 0.0001 in both cases) of the acute blood stasis experiments. As discussed above, group 1 and group 2 rats were exposed to both blood flow and acute blood stasis treatments, although the treatment order was reversed. In these groups, the acute blood stasis treatments were followed by lower mean I_i values compared with those obtained after blood flow treatments. These findings are consistent with the hypothesis that acute blood stasis reduces the interstitial uptake of albumin. Furthermore, the results suggest that the acute blood stasis effect was independent of treatment ordering and that the acute blood stasis effect was reversed by resuming blood flow.

Unlike the group 1 and group 2 experiments, the differences between the mean I_i values after the two treatments failed to reach statistical significance in group 3 rats (P = 0.4524), which received two consecutive flow treatments and no blood stasis treatment. This finding supports the idea that the acute stasis effect on interstitial albumin uptake is real and not the result of exposing and surgically manipulating the intestine or the result of consecutive BSA injections.

Perhaps the differences in interstitial albumin uptake following acute blood stasis, compared with normal flow, were due to differences in microvascular hydrostatic pressure. However, pressure measurements obtained at the site of cannulation, during the first and second fluorescent BSA uptake measurements, were not significantly different from each other (paired Student's t-test, P > 0.05). In 13 experiments involving all groups, mean pressures after the first and second treatments were 13 ± 4 and 13 ± 3 mmHg, respectively. Because the cannula was measuring pressure in the intestinal circulation when both vascular clamps were in place (preventing inlet and outlet flow and creating essentially a closed system), pressure was likely to be the same throughout the microvasculature. Thus the differences in interstitial albumin uptake cannot easily be explained by pressure differences.

Furthermore, the reduced interstitial albumin uptake after acute blood stasis cannot easily be accounted for by reduced capillary surface area available for albumin exchange, which is caused by the presence of RBCs that may have collected in vessels following the fluorescent BSA injection or due to venous backflow. With the use of the Optronics camera, the mean pooled I_i values were 182 ± 14 AIU (means ± SD; N = 12, where N is the number of interstitial areas measured) outside vessels containing RBCs compared with 179 ± 14 (means ± SD, N = 16) outside capillaries free of RBCs after stasis treatments.

Figure 4 shows plots obtained from grouping mean I_i values by time intervals after injection of fluorescent BSA. Data from group 1 and group 2 rats (Fig. 4, A and B, respectively) show that acute blood stasis treatments were followed by reduced mean I_i values in comparison to values obtained after normal blood flow. Data obtained after two consecutive flow treatments in group 3 rats show more similar I_i values pooled by time. Additionally, these data show little variation after 1 or 2 min, indicating that fluorescent BSA quickly reaches a steady state in the interstitium.

View larger version (15K): [in this window] [in a new window]   Fig. 4.   Mean interstitial fluorescence intensities (in arbitrary intensity units) in acute blood stasis experiments versus time. Capillary intensities and Ii values were pooled at selected time intervals for treatments 1 and 2 in group 1 rats, which had blood flow precede an acute blood stasis treatment (A), group 2 rats, which had acute blood stasis precede a normal blood flow treatment (B), and group 3 rats, which had two blood flow treatments (C).

Saline Stasis and Plasma COP-Equivalent Saline Stasis Experiments

	DISCUSSION

Top Abstract Introduction Materials & Methods Results Discussion References

The purpose of this study was to test the hypothesis that acute blood stasis in the intestine reduces the interstitial uptake of albumin from mucosal capillaries. We find that injection of fluorescent BSA into the intestinal circulation after blood flow and after acute blood stasis results in two clearly distinct patterns of fluorescence. A very diffuse pattern of fluorescence appears in the interstitial areas outside the capillaries if fluorescent BSA is intravascularly injected following brisk blood flow through the intestine. This pattern is consistent with rapid interstitial uptake of albumin. However, after 10-min flow stoppage, the capillaries brightly fluoresce, while surrounding interstitial areas retain darkness, producing high-contrast images. This suggests that blood flow stoppage prevented the rapid interstitial uptake albumin that was seen after normal flow.

Measurements of interstitial fluorescence from video recordings made after intravascular injection of fluorescent BSA show reduced values after acute blood stasis compared with those obtained after normal flow conditions. High interstitial fluorescence after blood flow and low interstitial fluorescence after acute blood stasis occur regardless of using FITC-BSA or TRITC-BSA as tracers and whether the acute blood stasis treatment precedes or follows the flow treatment. In experiments in which interstitial albumin uptake is measured after two consecutive flow treatments (group 3 rats), there is no difference between the measured interstitial fluorescence intensities. These findings support our hypothesis that acute blood stasis reduces the interstitial uptake of albumin in the intestinal mucosa. To our knowledge, this observation has not been reported previously.

There are, however, a number of possible explanations for observing reduced interstitial fluorescence after acute blood stasis. Possibilities include 1) stasis-induced edema obscuring fluorescence intensity; 2) reduction in available capillary surface area due to the presence of stationary cells within capillaries after the stasis treatments; 3) an adenosine 3',5'-cyclic monophosphate (cAMP) effect occurring during stasis related to isoproterenol in the suffusing solution; 4) tissue hypoxia during stasis; 5) release of vasoactive substances during stasis; 6) alterations in Starling forces, and hence convective flux, or lymph flow after stasis; and 7) changes in the transport properties of capillaries.

First, it seems unlikely that the reduced interstitial fluorescence intensity, which we observed after acute blood stasis, was due to tissue edema. We did not observe a reduction in interstitial fluorscence after plasma COP-equivalent saline stasis. Also, the microvessels were as easy to resolve by fluorescence microscopy after stasis as they were after normal blood flow. From other studies, we know that if the intestinal mucosa is made edematous by hemodilution, the vessels are difficult to bring to clear focus because they are obscured by the surrounding swollen tissue.

Second, differences in interstitial fluorescence intensity are not likely due to reductions in available capillary surface area following stasis. In our protocol, RBCs are flushed from the intestinal circulation before injection of fluorescent BSA. Failure to remove all RBCs in capillary lumens following the HBS flushes was rare. Reappearance of RBCs in the capillaries due to retrograde passage past the HPV clamp was also uncommon. In addition, reappearance of RBCs when the tracers were present in the vasculature was no more frequent after blood stasis than after normal flow. Nevertheless, when RBC-containing vessels were specifically sought and video recorded, I_i values outside of RBC-free capillaries exposed to stasis were similar to values obtained when RBCs were identified.

Another possible cause for the blood stasis effect on interstitial fluorescence relates to the presence of isoproterenol in the suffusion solution. Isoproterenol, apart from reducing peristaltic movement, increases intracellular cAMP (9). Increased cAMP levels have been reported to cause endothelial cell spreading, which has been associated with reductions in the permeability properties of cultured cells (12). It was possible that the blood stasis treatment might have reduced "wash out" of the isoproterenol applied in the suffusate, thereby increasing intracellular cAMP levels in the endothelium above those reached when flow was normal. However, in the saline stasis experiments, the HBS stasis treatment did not reduce tissue uptake of fluorescent BSA. During the saline stasis treatment, HBS containing isoproterenol was suffused at a rate of ~0.2 ml/min over the intestinal segment under observation, as occurred during all other experiments. Because the associated interstitial fluorescence intensity measurements were no different from those obtained after blood flow in these animals, the results are not supportive of isoproterenol causing the reduction in interstitial fluorescence after acute blood stasis.

The results of the saline stasis experiments and plasma COP-equivalent saline stasis experiments also argue against the acute blood stasis effect being the result of either tissue hypoxia or release of vasoactive substances due to clamped conditions. The stasis treatments in these experiments were not associated with lower interstitial fluorescence values.

Changes in Starling forces and lymph flow rate are very reasonable explanations for the differences in interstitial fluorescence that we observed after blood flow and after blood stasis. However, pressure measurements obtained from the aorta before and after stasis were similar. Pressure readings were taken when the inlet and outlet to the intestinal circulation were clamped, producing a closed system; thus the aortic pressure measurements should be similar to those within the capillaries. Although there was sometimes a reappearance of RBCs in the capillaries when the inlet and outlet were both clamped, any existing pressure differential within the system must have been extremely small as demonstrated by the very slow movement of the cells. More importantly, 10-min stasis conditions, when blood was replaced by albumin-containing HBS, were not followed by lowered interstitial fluorescent intensity measurements, as occurred after acute blood stasis. The Starling forces during acute blood stasis should match those created by using plasma COP-equivalent saline under clamped conditions. Because we did not measure lymph flow, we do not know if acute blood stasis alters lymph flow as a mechanism reducing interstitial fluorescence. Again, however, if this were the case, we would anticipate observing reduced interstitial fluorescence after stasis when plasma COP-equivalent saline is substituted for blood.

Finally, the differences in interstitial albumin uptake before and after acute blood stasis may be due to changes in the transport properties of the intestinal mucosal microvasculature. Reduced interstitial albumin uptake following acute blood stasis is consistent with our previous work suggesting that acute stasis causes the expression of a luminal negative surface charge (3), reduces the amount of intravascularly injected native ferritin entering the interstitium (2), and reduces the density of fenestrae in mucosal capillaries (8). Additionally, the group 2 acute blood stasis experiments show that the stasis effect is reversible. This result is consistent with our previous studies showing reversal of the acute stasis effect on endothelial surface charge in intestinal capillaries (3) and is consistent with the work of others on capillaries in skeletal muscle (10) and the hamster cheek pouch (11). Thus it appears that the factors reducing interstitial albumin uptake after acute blood stasis, be they electrostatic, structural, or due to combined causes, can be reversed, and it is unlikely that fixed, pathological changes occur after acute blood stasis in these capillaries. Furthermore, since after consecutive flow periods (group 3 rats) interstitial fluorescence was high, it appears that these vessels normally sustain a high albumin flux and that interruptions to flow produce reduced flux. Of interest, we have learned from additional work that the injection of fluorescent BSA while blood is flowing produces a diffuse fluorescent image similar to that seen following normal flow when fluorescent BSA is held in the intestinal circulation by clamps.

In addition to the stasis effect on interstitial albumin uptake, a final observation made during our studies warrants further comment. Plots of pooled I_i values versus time for a given experimental group showed little variation after 1 or 2 min, indicating that interstitial albumin quickly reaches a steady state. This is consistent with findings of Brooks and Dobbins (4), who demonstrated the presence of albumin in the interstitium and central lacteal of intestinal villi 1 min after intravenous injection and little variation in albumin concentration between 1 and 30 min after injection. A rapid rise in albumin uptake is also consistent with the work of Allan and Trier (1), who showed that intravenously injected horseradish peroxidase appeared in the pericapillary space of 97% of villus capillaries only 2 min after injection. It is not surprising that injected albumin reaches a steady state so quickly, since there is very little distance between the capillary network and the central lacteal in each villus.

Because this work is limited to the capillaries of the intestinal mucosa, we do not know if the response to acute blood stasis described here is a general microcirculatory reaction to stationary blood or is unique to mucosal capillaries; after all, mucosal capillaries are fenestrated, containing pathways not present in the majority of microvessels. Furthermore, it may be that the response to stasis is related to normal intestinal physiology. The intestine is typically prone to wide fluctuations in blood flow throughout the day: after meals, blood flow to the intestine is high, whereas in times of fast (or exercise), flow to the intestine is lowered. In the intestinal mucosa, capillaries may experience acute stasis physiologically and respond by mechanisms to reduce interstitial albumin uptake. A possible physiological advantage is that stasis-induced albumin retention in the vasculature would prevent excessive albumin concentrations outside capillaries when its transport functions are not needed. Additionally, albumin retention in the vascular space would assist in reabsorption of water into the circulation and thus in maintenance of blood volume. Future studies will be designed to address these issues.