FLOW MOTION IN THE INTESTINAL VILLI DURING HEMORRHAGIC... : Shock (original) (raw)
INTRODUCTION
The major purpose of cardiovascular resuscitation is restoration of the energy content of the tissues through normalization of the oxygen delivery. In hypovolemic hemorrhagic shock (HS), this could ideally be achieved by the reestablishment of precapillary pressure to preshock values and maintenance of the surface area needed for the exchange of nutrients and drainage of waste products. Restoration of microcirculation is a good indicator of the effectiveness of resuscitation strategies, but commonly used parameters do not allow a comparison of dynamically varying microcirculatory changes. Specifically, the reduction of circulating blood volume leads to peculiar periodic oscillations termed flow motion in capillaries of several organs, including pancreas, brain, and skeletal muscle (1–3). To date, only a rough estimate could be given to characterize the fluctuations in time (i.e., cycles per minute), and, therefore, the comparison of continuous and pulsatile perfusion states (or two pulsatile states) is difficult. Similarly, studies that have relied on blood flow or velocity values invariably ignored the distinction between steady and nonsteady (fast/slow flow transition) states during their comparisons. The main goal of our study was to address this problem by providing an approach to more accurately quantify the time-wise heterogeneity of tissue microcirculation during compromised perfusion. In this respect, the response of small intestinal microcirculation is of special interest because the intestinal mucosa is critically affected by HS-induced systemic reflex vasoconstriction and flow redistribution (4). To this end, intravital videomicroscopy with the orthogonal polarization spectral (OPS) imaging technique was used for the continuous detection of the microvascular alterations in the intestinal villi in a large-animal model of HS. This novel diagnostic method uses reflected polarized light to visualize hemoglobin-containing structures without the additional use of a fluorochrome (5–7). Furthermore, we have introduced a new probabilistic-based method for the accurate characterization of the statistical properties of flow motion.
Second, we assessed the efficacy of various resuscitation fluids as compared with physiological saline. Hypertonic-hyperoncotic solutions evoke a rapid transcapillary fluid shift from the intracellular and interstitial space to the intravascular compartment (8). As a result, stabilization of macrohemodynamics, improvement of the microcirculation, and, eventually, an improved outcome can be expected (9–12). Similarly, there is a growing body of evidence that endothelin-1 (ET-1) peptide and the activation of vasoconstrictor ET receptors play decisive roles in acute microcirculatory disorders of the cardiovascular system (13). It has been demonstrated that plasma ET-1 levels linearly correlated with the amount of blood loss during HS (14, 15) and ET-A receptor antagonism reduce intestinal microvascular and tissue injury, and polymorphonuclear leukocyte accumulation during ischemia-reperfusion (16, 17). Therefore, we have examined the combined effects of hypertonic-hyperoncotic solutions and ET-A receptor antagonism on HS-induced microcirculatory reaction in the canine small intestine.
MATERIALS AND METHODS
The experiments were performed in accordance with the National Institutes of Health guidelines and with the approval of the Ethical Committee for the Protection of Animals in Scientific Research of the University of Szeged. The experiments were performed on a total of 32 mongrel dogs (15.9 ± 0.4 kg) under intravenous sodium pentobarbital anesthesia (30 mg/kg).
The left femoral artery and vein were cannulated for measurement of the mean arterial pressure (MAP) and for fluid and drug administration, respectively. A Swan-Ganz catheter (Braun Melsungen, AG, Melsungen, Germany) was introduced into the pulmonary artery through the right femoral artery for intermittent measurement of the cardiac output (CO). Blood withdrawal was performed through a catheter inserted into the right femoral artery.
The animals were mechanically ventilated, and the respiratory rate and tidal volume were adjusted to maintain the arterial pCO2 at 40 ± 5 torr (FiO2 33%). The core temperature was maintained at 39°C ± 0.5°C. After a midline laparotomy, an ultrasonic flow probe (Transonic Systems, Ithaca, NY) was placed around the artery supplying the terminal ileum. A small enterotomy was performed and a tonometric balloon catheter (Instrumentarium Corp., Helsinki, Finland) was introduced into the lumen to measure intramucosal pH (pHi). Proximal to this enterotomy, the bowel was opened along the antimesenteric border, placed on pedestal, and the mucosal microcirculation was visualized with the OPS imaging technique. The exteriorized segment was continuously superfused with warm saline solution and was covered with plastic wrap to avoid heat loss and evaporation. The animals received physiologic saline infusion at a rate of 15 mL/kg/h during the surgical procedure.
Experimental protocol
The animals were randomly assigned to four groups. Group 1 served as sham-operated control (n = 3), and groups 2 through 4 were subjected to HS. Surgery was followed by a 60-min stabilization period. Baseline variables were recorded for 30 min, and then blood was withdrawn from the femoral artery into a heparinized (25 IU/mL) reservoir until the MAP reached 40 mmHg. Blood was additionally withdrawn or retransfused to maintain the set MAP value for 60 min. The animals were monitored for 180 min after HS and were then killed with an overdose of pentobarbital.
In Group 2 (n = 11), the animals were resuscitated with 0.9% saline (150% of the lost blood volume) over 10 min, followed by a low-rate infusion of saline (1 mL/kg/h). Group 3 (n = 10) was treated with hypertonic saline-Dextran (HSD) solution (prepared from 10% isotonic Dextran-40; Baxter, Muenchen, Germany, and 7.2% NaCl solution) 4 mL/kg over 10 min, followed by a continuous infusion of saline (1 mL/kg/h). The animals in Group 4 (n = 8) were treated with the selective ET-A receptor antagonist ETR-p1/fl peptide (18) (VLNLCALSVDRYRAVASWRVI; Kurabo Ltd., Osaka, Japan; 100 nmol/kg i.v.. bolus in 1.5 mL/kg saline) 5 min before resuscitation, and HSD (4 mL/kg over 10 min) and saline infusion (1 mL/kg/h) were then given. The rates and volumes of blood loss were not statistically different between the groups (42.5 ± 2.2 mL/kg, 42.5 ± 1.8 mL/kg, and 44.4 ± 2.0 mL/kg in saline-, HSD-, and HSD+ETR-p1/fl-treated groups, respectively).
Measurements of macro- and microhemodynamics, systemic and pulmonary arterial pH, pO2, pCO2, hemoglobin, and O2 saturation were performed at baseline and then in 15, 30-, or 60-min intervals, respectively. pHi was calculated by the modified Henderson-Hasselbach formula with a correction factor for 30-min equilibration. MAP, pulmonary arterial pressure (PAP), and central venous pressures were measured with Statham P23Db transducers. Pressure and blood flow signals were continuously monitored and registered with a computerized data-acquisition system (Hemosys 1.17; Experimetria Ltd., Budapest, Hungary). CO was determined by thermodilution using a Cardiostar CO-100 computer (Experimetria Ltd.,) normalized for body weight and expressed as cardiac index (CI, mL/kg/min). Arterial blood gases, fractional oxyhemoglobin saturation, and intramucosal pCO2 were measured with a blood-gas analyzer (AVL, Graz, Austria).
Arterial O2 content was calculated as hemoglobin concentration × 1.36 × arterial O2 saturation + 0.0031 × arterial O2 tension. Systemic O2 consumption was calculated as CI multiplied by the difference between arterial and venous O2 content.
The intravital OPS technique (Cytoscan A/R; Cytometrics, Philadelphia, PA) with a 10× objective was used for continuous visualization of the microcirculation of the intestinal villi (Fig. 1). This technique uses reflected polarized light at the wavelength of the isobestic point of oxy- and deoxyhemoglobin (548 nm). Because polarization is preserved in reflection, only photons scattered from a 2- to 300-μm depth contribute to the image formation (5). Images were recorded by an S-VHS video recorder (AG-TL 700; Panasonic, Secaucus, NJ) and were evaluated offline by frame-to-frame analysis. The functional capillary density (FCD, length of perfused nutritive capillaries per observation area (cm−1) and the red blood cell velocity (RBCV, μm/s) were determined in three separate fields by means of a computer-assisted image analysis system (IVM Pictron, Budapest, Hungary). During capillary flow motion, RBCV was determined during high-flow and low-flow (or stop) conditions. All data are expressed as the means of a minimum of 10 measurements at each time point.
Representative micrograph of canine intestinal mucosal villi, obtained with the OPS technique.
Statistical analysis of RBCV
Microcirculatory flow was continuous under control conditions in the villi, whereas cyclic fluctuation appeared during HS (Fig. 2). During oscillatory flow, calculation of the average RBCV should not be performed by using the arithmetic mean of individual RBCV values. There are several important factors that should carefully be considered. First, the RBC flow displays an alternating sequence of high-velocity (V) and low-velocity (v) pattern. Because the transition from a high-velocity period to a low-velocity period and vice versa always occurs abruptly within less than a second, the time course of the flow is similar to a square wave and not a sinusoidal wave. Second, after the HS, the microcirculatory flow pattern displayed a remarkably regular behavior in each experiment within the recorded observation period. Third, if certain measured values of the velocity have considerably longer durations than others, then the duration of the velocity should also be taken into account in the calculation of the average RBCV. In addition to the velocity itself, the duration of the individual high- or low-flow periods can also be accurately measured, therefore, the above oscillatory pattern can be conveniently described mathematically using probabilities as follows. If the nth value of the high velocity component Vn is observed for a time interval Tn, then the likelihood of finding Vn during the total measurement time of high flows (TH) can be taken to be proportional to Tn. In other words, the probability Pn of observing Vn can be estimated as the ratio Pn = Tn/TH. The set of Pn values as a function of Vn defines the probability distribution of V, which fully characterizes the random variable V and allows us to calculate its statistical properties in a rigorous manner. For example, the expected or average value (Vave) of the high-velocity V is the first moment of the distribution:
Schematic diagram of hypoperfusion-induced villus capillary flow motion. Time course of RBCV changes in a representative experiment (A). Calculations of cycles, relative duration of high-flow periods, and probabilities are shown. A-RBCV is calculated as the average of the distribution (Eq. 3). In B, A-RBCV is calculated as the weighted average of the high- and low-velocity averages.
where N is the total number of high-flow observations. The variance of V can be calculated from the second moment of the distribution as
The SD and SE can be easily obtained from Equation 2 by taking the square root and by scaling Equation 2 with (n − 1)/n before taking the square root, respectively. A similar distribution can be constructed for the low-velocity v by calculating the probabilities pm = tm/TL, where tm is the observed duration of the mth observed value vm and TL is the total duration of low-flow periods. The expected value and the variance of v are calculated similarly to those of V by replacing Pn and Vn by pm and vm, respectively, and summing over the total number of observations of low flow periods (M). Finally, the statistical properties of RBCV from the combined low-flow and high-flow data are obtained by first noting that the total measurement time is now T = TH + TL. Next, we recalculate the probabilities as Pn = Tn/T and pm = tm/T and take the first moment of the combined distribution as:
The SD is the square root of the variance given by:
If these distributions are reasonably uniform, the probabilities Pn and pm can be approximated as 1/N and 1/M, respectively. Thus, we can simply replace the calculations of Vave and vave by the simple arithmetic means of the measured V and v values, respectively. However, in general, the total durations of the high and low velocities are different, and, therefore, one cannot use the arithmetic mean to calculate the average RBCV. In this case, Equation 3 reduces to the weighted sum of Vave and vave:
where T = TH + TL. Similarly, the variance of RBCV is given by:
Whether Equations 5 and 6 can be used instead of the more general Equations 3 and 4 depends on the shape of the distributions P and p. Because the individual distributions may vary from animal to animal, we chose to test the uniformity of these distributions by correlating the average and SD of RBCV for each animal obtained from Equations 5 and 6 with those obtained from Equations 3 and 4. Figure 2 shows these calculation schemes.
Serum endothelin level measurements
Blood samples (5 mL) were obtained at the beginning of the experiments in the 60th min of HS, and in the 15th, 60th, and 180th min of resuscitation. The samples were analyzed for ET-1 with an ELISA kit (Biomedica, Vienna, Austria).
Myeloperoxidase (MPO) enzyme activity measurements
The MPO activity, as an index of the neutrophil accumulation in the tissues, was determined by using a modified method of Kuebler et al. (16, 19).
Statistical analysis
Data are expressed as mean ± SEM. Data analysis was performed with a statistical software package (SigmaStat for Windows; Jandel Scientific, Erkrath, Germany). Changes in variables within groups were analyzed by two-way analysis of variance tests followed by the Bonferroni’s test. Differences between groups were evaluated by means of Student’s unpaired t test. P values < 0.05 were considered significant.
RESULTS
In the sham-operated group, the macro- and microhemodynamic parameters did not change significantly during the 330-min observation period (data not shown).
The approximate 50% reduction of the calculated blood volume was accompanied by an approximate 70% decrease in CI (Fig. 3, A and B). Resuscitation was followed by a partial recovery in MAP, irrespective of the used therapy. Resuscitation with saline resulted in a short-term restoration of CI at the onset of resuscitation, but this was followed by a gradual decline. Similar macrohemodynamic deterioration was observed after HSD treatment. However, ET-A antagonisms resulted in a lesser degree of recovery in CI as compared with saline or HSD treatment at the onset of reperfusion (Fig. 3B). The ET-A antagonist treatment did not influence the heart rate (data not shown).
Macrohemodynamic changes during 60-min HS and 180-min resuscitation. MAP (A), CI (B), Sflow (C), and PAP (D) in saline-, HSD-, and HSD + endothelin-A receptor antagonist-treated groups. Data are means ± SEM.
The intestinal perfusion was reduced by approximtely 70% during HS. Although the blood flow exceeded the baseline after the start of resuscitation with saline, a gradual deterioration was observed, parallel to CI changes (Fig. 3B) and no difference was detected between the experimental groups after 15 min.
PAP was decreased by 30% to 50% during HS and then returned to baseline during the early phase of resuscitation (Fig. 3D). ET-A receptor antagonism induced a gradual decrease, which was statistically different from other treatments in the late phase of resuscitation.
HS induced a 4-fold increase in O2 extraction, and persistent systemic metabolic acidosis. The intestinal pHi was also diminished until the end of the experiments in all groups, demonstrating indirectly the impaired local tissue perfusion. HSD alone or in combination with ET-A receptor antagonist ameliorated the acidosis to some extent, as evidenced by improved base excess and bicarbonate levels during the later phase of resuscitation (Table 1).
Effect of saline, HSD, or HSD + endothelin-A receptor antagonism on various blood gas parameters, hemoglobin concentration, and intramucosal pH in dogs subjected to HS
Microcirculatory flow was continuous at the villus tips under control conditions, whereas cyclic fluctuation appeared during HS. This time-dependent flow motion was not confined to the capillaries—it could be observed in subepithelial venules and central arterioles as well. Additionally, alternating (on-off) flow evolved within adjoining villi, i.e., spatially and temporally synchronized flow periods were observed. Figure 4, A and B demonstrates that there is a good correlation between average RBCV computed from Equations 3 and 5 and the SD values computed from Equations 4 and 6. This implies that the distributions of the high and low velocities during flow motion are reasonably uniform, hence, the simplified weighted averages (Eqs. 5 and 6) can be used to accurately characterize flow motion. In contrast, Figure 4, C and D suggests that simple arithmetic averaging is inappropriate to characterize the average RBCV and its SD. Using Equation 5, the A-RBCV was 403 ± 67 μm/s, whereas the simple arithmetic mean was 341 ± 33 μm/s (P < 0.001). More importantly, the SD obtained from simple arithmetic calculations did not even correlate with the SD obtained from Equation 4, and the values were significantly overestimated (P < 0.001). Comparing the population means of RBCV before and after HS resulted in a statistically significant reduction in the average values. The SD values slightly decreased (using Eq. 6) from 102 ± 40 μm/s in baseline to 87 ± 29 μm/s in HS, but the SD based on arithmetic mean increased significantly to 254 ± 46 μm/s (P < 0.001), again suggesting its inappropriate use during flow motion. Although the SD decreased, the coefficient of variation actually increased from 0.15 ± 0.06 to 0.22 ± 0.08, which was significant (P < 0.01). The coefficient of variation based on arithmetic mean (0.74 ± 0.10) was again significantly overestimated (P < 0.001).
Correlations between methods estimating A-RBCV and its SD taken from representative calculations at 30 min of HS. A and B show A-RBCV and its SD from Eqs. 3 and 4, respectively, compared with those obtained from Eqs. 5 and 6, respectively. The correlation coefficients are high and the slope of the regression line is nearly unity. C and D show the correlations of A-RBCV and its SD calculated via arithmetic averaging and from Eqs. 3 and 4, respectively.
During HS, periods of high RBCV (∼500-600 μm/s, Fig. 5A) were followed by periods of low RBCV (∼100-150 μm/s). The average duration of high- and low-flow periods was 9.9 ± 0.4 s and 7.4 ± 0.5 s, respectively. For a 60-s interval, this equals 3.5 cycles/min, or corresponds to an approximate 43% decrease in relative duration of high-flow periods as compared with the continuous baseline flow (100%). At the onset of resuscitation, continuous flow periods were transiently seen in 33%, 40%, and 50% of the experiments after saline, HSD, and HSD + ETR p1/fl treatment, respectively. During the later stages of resuscitation, the relative duration of high RBCV periods was decreased, indicating the predominance of oscillatory flow in the villi (Fig. 5B).
Microhemodynamic changes during 60-min HS and 180-min resuscitation. Changes RBCV during high flow in the capillaries (A), relative duration of high-flow periods (B), A-RBCV (C), and FCD (D) in saline-, HSD-, and HSD + endothelin-A receptor antagonist-treated groups. Data are means ± SEM.
The ET-A receptor antagonism significantly increased the relative duration of the high RBCV periods at the onset of resuscitation either by prolonging the continuous flow or the duration of high-flow periods during oscillation. In the event of prolonged high-flow periods, the use of cycles/minute would be particularly misleading to express the periodicity because it would underestimate, or mask, these favorable alterations (Table 2).
Comparison of data obtained from the calculation of relative duration of high-flow periods with those for the cycles. Data on the 60 min of resuscitation (t = 120 min) are shown
During HS, the RBCV during the high-flow phases was significantly lower than at the baseline (Fig. 5A), and returned to the control level in all groups at the onset of resuscitation. However, the RBCV was moderately higher in the HSD- and HSD + ETR-p1/fl-treated groups during the high-flow periods. A-RBCV, which allows simultaneous consideration of RBCV and the flow pattern changes, was restored at the onset of resuscitation in response to HSD and HSD + ETR p1/fl treatments (Fig. 5C).
Capillary stasis in hemoglobin-containing structures was not observed during HS. Resuscitation was associated with a decrease in FCD in each group, and a return to control level was observed in HSD and HSD + ETR-p1/fl-treated groups only (Fig. 5D).
Plasma ET-1 was significantly elevated by the end of the shock period in each group (from 3.93 pg/mL ± 0.52 to 6.77 ± 0.5 pg/mL; P = 0.0016). At the onset of resuscitation, the high level of ET-1 persisted in both HSD-treated groups, however, a decrease occurred in response to resuscitation with saline (to 2.87 ± 0.71 pg/mL; P = 0.353 vs. baseline; P < 0.05 vs. HSD groups). These changes were followed by a gradual increase until the end of the examination period in each group reaching 8.95 ± 2.86 pg/mL maximal values in saline-, 12.84 ± 1.68 pg/mL in HSD-, and 12.12 ± 1.94 pg/mL in HSD + ETR p1/fl-treated groups, respectively.
HS followed by 180 min of resuscitation was accompanied by a 54.2%, 116.4%, and 52.4% increase in MPO activity in the intestine in animals resuscitated with saline, HSD, and HSD combined with the ET-A receptor antagonist, respectively (from 2.25 ± 0.25 to 3.47 ± 0.43, 4.87 ± 0.82; P < 0.05 vs. baseline and 3.43 ± 0.72 U/mg protein, respectively).
DISCUSSION
In this study, the intestinal macro- and microcirculatory consequences of 50% blood volume reduction and the effects of low-volume resuscitation were investigated. The severity of hemorrhage was marked by the approximate 70% decreases in CO and ileal blood flow. Resuscitation caused a partial and transient restoration in macrohemodynamics that was followed by a gradual deterioration irrespective of the applied therapy. Nevertheless, hypertonic solutions induced significant microcirculatory improvement in the early phase of resuscitation, marked by continuous flow pattern and normalization of RBCV in the intestinal villi. However, these microcirculatory alterations could be characterized only by means of a novel method of data analysis.
Similar to other organs such as pancreas, brain, skeletal muscle, and skin (1–3, 20–21), the deteriorating intestinal macrocirculation was accompanied by periodic fluctuations in capillary blood flow. This occurred not only during the shock state, but also during the later phase of resuscitation, when the gradually decreasing CI was accompanied by diminished ileal blood flow. Because this microvascular oscillatory phenomenon was not observed under control conditions, its appearance can be considered a manifestation of tissue malperfusion at any later stage of the experiments. Thus, it appears that the beneficial effect of fluid resuscitation is related not only to the temporarily restoration of near-normal microcirculatory velocity values, but also to the re-established continuous flow conditions. Accordingly, the increased length of high-flow phases during flow motion also reflects improved tissue perfusion. Therefore, the duration of distinct flow periods is a critical factor and should be taken into account in the data analysis and interpretation.
It has been recognized that maintenance of an adequate microvascular blood supply is critical for the survival and function of the reperfused tissues. Certain microcirculatory changes such as altered RBCV can be quantified by conventional methods. However, when a variable is dynamically changing over time, generally used descriptive parameters, e.g., the simplistic approach of calculating cycles, can be misleading. First, cycles do not allow for the quantification of high flow periods within cycles. Second, the use of cycles as a measure of variability is not applicable when perfusion becomes continuous, even for a limited part of the experiments.
Because of the changes not only in the magnitude, but also in the duration of individual flow periods over time, a comparison with baseline conditions (or steady states with oscillatory blood flow), and the evaluation of changes in distinct experimental groups (characterized by different flow patterns at a given time point) is very difficult. In addition, the alternating flow conditions in adjoining structures can also be present within a given time frame. These considerations led us to seek a new method to quantify the average microcirculatory flow and its fluctuations, which takes both the changes in amplitude and duration simultaneously into account. Using a probabilistic approach, the instantaneous velocity was considered as a random variable and the duration for which a particular value of the velocity was observed was used to calculate the probabilities as the relative duration of the high- and low-oscillatory flow periods.
Because the simple weighed arithmetic average of RBCV (Eq. 5) and its SD (from Eq. 6) correlated well with the full probabilistic approach (Eqs. 3 and 4, respectively), the former calculations proved to be a simple and reproducible method for characterizing the statistical features of flow motion. This detailed data analysis could demonstrate that the average RBCV decreased and its coefficient of variation increased during HS. More importantly, the analysis could reveal that hyperoncotic solution supplemented with an ET-A receptor antagonist exerted its beneficial effect by maintaining continuous flow in the villi and prolonging the length of high-flow periods at the onset of resuscitation.
According to the present view, flow motion develops on the basis of vasomotion, i.e., “rhythmic oscillations superimposed on tonic contractions” (see Ref. 20 for a review). A causal relationship between these microvascular phenomena was proposed by Vollmar et al. (1, 23) on the basis of their simultaneous occurrence as seen in the skeletal muscle and pancreas. It is interesting that the existence of vasomotion was proven in most downstream arterioles under physiological conditions, for instance, in the muscle, brain, and mesenteric microcirculation (24–26), but the presence of flow motion at the capillary level could not be revealed by standard videomicroscopic methods (23). We could not demonstrate capillary flow motion under normal circumstances either. Therefore, it is conceivable that the oscillatory changes in the mucosal oxygenation reported by others are possibly not due to significant baseline capillary flow motion (27). It is reasonable to presume, however, that even if present, physiologic capillary flow motion has small amplitude and cannot be detected by standard frame to frame analysis of the recordings. In vitro data may support this view because transmural pressure and amplitude of rhythmic contractions are found to be negatively correlated (28). Another interpretation of this problem would be to state that hypovolemic hypotension increases the amplitude of vaso- and flow motion, i.e., causes so great low/high value differences of RBCV that it can easily be traced during data analysis.
The cause of flow motion is unclear, but a new balance between vasoconstrictor and vasodilator forces can be presumed behind this phenomenon. Flow oscillations could improve the efficacy of tissue oxygenation during low-flow conditions (29). HS is characterized by an intense neuroendocrine response and stress hormone release (30). Norepinephrine induces periodic, intracellular calcium-mediated changes in the membrane potential of vascular smooth muscle cells in vitro (22), and increases flow oscillations in the brain (2). Therefore, it is conceivable that endogenous catecholamines contribute to local flow pattern regulation. Furthermore, it has been shown that endothelial ET-1 production is enhanced by epinephrine (31). This suggests that the evolving vasoconstriction may be amplified by ET-1 release during HS.
Vasodilator factors are equally important in regulating flow motion because vasomotion is dependent on an intact endothelium (28). It has been proposed that the interstitial accumulation of vasodilator metabolites leads to precapillary sphincter opening during HS (10). A role of CO2 can be presumed because inhaled CO2 reversibly abolishes flow oscillations (2). Therefore, it is likely that arteriolar vasoconstriction is periodically overcome by metabolic or endothelium-derived dilator forces. It is important that the oscillatory phenomenon was not confined to the capillaries: the cessation of flow started in the central arteriole. Hence, vasoconstriction did not arise from precapillary sphincters, but probably from 3A-type upstream arterioles (according to the description of Bohlen and Gore [32]). The synchronized flow patterns of adjoining villi support this hypothesis.
The circulating levels of ET-1 were significantly increased during HS and peaked at the end of the observation period. At the onset of resuscitation, significantly higher ET-1 levels could be demonstrated in both HSD-treated groups with or without ET-A antagonism than after saline treatment. This is in agreement with the original report, whereby the ET-A receptor inhibitor compound exerted its inhibitory effects directly through specific receptor binding rather than influencing the circulating levels of endogenous ET-1 (18). ET-1 is the most powerful vasoconstrictor substance known to date and the vasoconstrictive effects are mediated predominantly via the ET-A receptors present on the vascular smooth muscle cells. Endogenous ET enhances cardiac contractility (33), and ET-A receptor antagonism transiently deteriorated the cardiac performance. Besides, the prolonged effect of a bolus dose of ET-A antagonist on the pulmonary artery pressure underlines the role of ET-1 in the regulation of pulmonary vascular tone. The beneficial effect of ET-A antagonism on the villus microcirculation was also evidenced by the reduced periods of flow cessation. The microcirculatory effects of endogenous ET can be linked to a calcium-dependent signal transduction mechanism. ET-1 causes sustained oscillatory elevations in intracellular calcium level in vitro, which could be prevented by ET-A receptor antagonism in vivo (34). Thus, the ET-induced calcium-mediated catecholamine release could modify microcirculatory responses through an ET-A receptor-dependent mechanism (35).
A role for hemorrhage-induced oscillations in preventing a reduction in FCD has been proposed and calcium channel blockage simultaneously reduces flow motion and FCD (23). FCD is defined as the length of red cell-perfused capillaries in relation to the observation area, which accurately describes the decrease in the efficacy of tissue perfusion when the corresponding area is unchanged (36). However, the measurement was established for the visualization of perfused capillaries with fluorescent markers. In our study, FCD virtually ranged between zero and maximal values, reflecting a mechanism causing nearly complete stasis in all structures of the villi for well-defined time periods. Edema or shrinkage of the reference area poses further problems. The intercapillary distance (inversely related to FCD) may increase after resuscitation, and if the capillary perfusion rate does not change, this leads to an underestimation of FCD. Because we did not observe changes in the capillary recruitment or perfusion rate in response to HSD, the decrease in FCD in response to resuscitation with saline may be the consequence of fluid movement into the interstitium. Edema can also cause compression of capillaries, and the simultaneous consideration of FCD and perfusion rate is therefore recommended.
In summary, in the presence of time varying factors, approaches should be emphasized that simultaneously consider all major components of variability. In this study, we introduced a probabilistic mathematical method that satisfies this requirement and provides a basis for comparison of different microcirculatory reactions in different organs. With these calculations, we could determine the microcirculatory improvement after hyperoncotic resuscitation and quantitatively compare the microvascular flow-related alterations due to ET-A receptor inhibition.
REFERENCES
1. Vollmar B, Preissler G, Menger MD: Hemorrhagic hypotension induces arteriolar vasomotion and intermittent capillary perfusion in rat pancreas. Am J Physiol 267:H1936–H1940, 1994.
2. Hudetz AG, Roman RJ, Harder DR: Spontaneous flow oscillations in the cerebral cortex during acute changes in mean arterial pressure. J Cereb Blood Flow Metab 12:491–499, 1992.
3. Borgstrom P, Schmidt JA, Bruttig SP, Intaglietta M, Arfors KE: Slow-wave flow motion in rabbit skeletal muscle after acute fixed-volume hemorrhage. Circ Shock 36:57–61, 1992.
4. Toung T, Reilly PM, Fuh KC, Ferris R, Bulkley GB: Mesenteric vasoconstriction in response to hemorrhagic shock. Shock 13:267–273, 2000.
5. Groner W, Winkelman JW, Harris AG, Ince C, Bouma GJ, Messmer K, Nadeau RG: Orthogonal polarization spectral imaging: a new method for study of the microcirculation. Nat Med 5:1209–1212, 1999.
6. Nadeau RR, Groner W: Ortogonal polarization spectral imaging: state of the art. In Messmer K (ed.): Progress in Applied Microcirculation. Vol 24. Basel, Switzerland: Karger, 2000, pp 9–20.
7. Tugtekin I, Theisen M, Matejovic M, Stehr A, Ploner F, Trager K, Mathura KR, Ince C, Georgieff M, Radermacher P: Endotoxin-induced ileal mucosal acidosis is associated with impaired villus microcirculation in pigs. In Messmer K (ed.): Progress in Applied Microcirculation. Vol 24. Basel, Switzerland: Karger, 2000, pp 61–69.
8. Frohlich ED: Prolonged local and systemic haemodynamic effects of hyperosmotic solutions. Arch Int Pharmacodyn 161:154–166, 1966.
9. Jonas J, Heimann A, Strecker U, Kempski O: Hypertonic/hyperoncotic resuscitation after intestinal superior mesenteric artery occlusion: early effects on circulation and intestinal reperfusion. Shock 14:24–29, 2000.
10. Dries DJ: Hypotensive resuscitation. Shock 6:311–316, 1996.
11. Victorino GP, Newton CR, Curran B: Dextran modulates microvascular permeability: effect in isotonic and hypertonic solutions. Shock 19:183–186, 2003.
12. Shi HP, Deitch EA, Da Xu Z, Lu Q, Hauser CJ: Hypertonic saline improves intestinal mucosa barrier function and lung injury after trauma-hemorrhagic shock. Shock 17:496–501, 2002.
13. Yanagisawa M, Kurihara H, Kimura S, Tomobe Y, Kobayashi M, Mitsui Y, Yazaki Y, Goto K, Masaki T: A novel potent vasoconstrictor peptide produced by vascular endothelial cells. Nature 332:411–415, 1988.
14. Chang H, Wu GJ, Wang SM, Hung CR: Plasma endothelin level changes during hemorrhagic shock. J Trauma 35:825–833, 1993.
15. Schlichting E, Aspelin T, Grotmol T, Lyberg T: Endothelin and hemodynamic responses to superior mesenteric artery occlusion shock and hemorrhagic shock in pigs. Shock 3:109–115, 1995.
16. Wolfárd A, Vangel R, Szalay L, Kaszaki J, Haulik L, Balogh A, Nagy S, Boros M: Endothelin-A receptor antagonism improves small bowel graft perfusion and structure following ischemia and reperfusion. Transplantation 68:1231–1238, 1999.
17. Massberg S, Boros M, Leiderer R, Baranyi L, Okada H, Messmer K: Endothelin (ET)-1 induced mucosal damage in the rat small intestine: role of ET(A) receptors. Shock 9:177–183, 1998.
18. Baranyi L, Campbell W, Ohshima K, Fujimoto S, Boros M, Okada H: The antisense homology box: a new motif within proteins that encodes biologically active peptides. Nat Med 1:894–901, 1995.
19. Kuebler WM, Abels C, Schuerer L, Goetz AE: Measurement of neutrophil content in brain and lung tissue by a modified myeloperoxidase assay. Int J Microcirc 16:89–97, 1996.
20. Chen CW, Hsiue TR, Chang HY: Effects of _N_ω-nitro-l-arginine (L-NOARG) on blood flow and vasomotion in rat diaphragm microcirculation during hemorrhagic hypotension. Shock 12:69–74, 1999.
21. de Carvalho H, Dorigo D, Bouskela E: Effects of Ringer-acetate and Ringer-Dextran solutions on the microcirculation after LPS challenge: observations in the hamster cheek pouch. Shock 15:157–162, 2001.
22. Gustafsson H: Vasomotion and underlying mechanisms in small arteries. An in vitro study of rat blood vessels. Acta Physiol Scand Suppl 614:1–44, 1993.
23. Rucker M, Strobel O, Vollmar B, Roesken F, Menger MD: Vasomotion in critically perfused muscle protects adjacent tissues from capillary perfusion failure. Am J Physiol 279:H550–H558, 2000.
24. Colantuoni A, Bertuglia S, Intaglietta M: Microvessel diameter changes during hemorrhagic shock in unanesthetized hamsters. Microvasc Res 30:133–142, 1985.
25. Hundley WG, Renaldo GJ, Levasseur JE, Kontos HA: Vasomotion in cerebral microcirculation of awake rabbits. Am J Physiol 254:H67–H71, 1988.
26. Minamiyama M, Hanai S: Propagation properties of vasomotion at terminal arterioles and precapillaries in the rabbit mesentery. Biorheology 28:275–286, 1991.
27. Hasibeder W, Germann R, Sparr H, Haisjackl M, Friesenecker B, Luz G, Pernthaler H, Pfaller K, Maurer H, Ennemoser O: Vasomotion induces regular major oscillations in jejunal mucosal tissue oxygenation. Am J Physiol 266:G978–G986, 1994.
28. Gustafsson H, Nilsson H: Rhythmic contractions of isolated pressurized small arteries from rat. Acta Physiol Scand 152:145–152, 1994.
29. Tsai AG, Intaglietta M: Evidence of flow motion induced changes in local tissue oxygenation. Int J Microcirc Clin Exp 12:75–88, 1993.
30. Lilly MP, Engeland WC, Gann DS: Adrenal medullary responses to repeated hemorrhage in conscious dogs. Am J Physiol 251:R1193–R1199, 1986.
31. Kohno M, Murakawa K, Yokokawa K, Yasunari K, Horio T, Kurihara N, Takeda T: Production of endothelin by cultured porcine endothelial cells: modulation by adrenaline. J Hypertens Suppl 7:S130–S131, 1989.
32. Bohlen HG, Gore RW: Preparation of rat intestinal muscle and mucosa for quantitative microcirculatory studies. Microvasc Res 11:103–110, 1976.
33. Watanabe T, Kusumoto K, Kitayoshi T, Shimamoto N: Positive inotropic and vasoconstrictive effects of endothelin-1 in in vivo and in vitro experiments: characteristics and the role of L-type calcium channels. J Cardiovasc Pharmacol 13:S108–S111, 1989.
34. Guibert C, Beech DJ: Positive and negative coupling of the endothelin ETA receptor to Ca2+-permeable channels in rabbit cerebral cortex arterioles. J Physiol 514:843–856, 1999.
35. Hinojosa-Laborde C, Lange DL: Endothelin regulation of adrenal function. Clin Exp Pharmacol Physiol 26:995–999, 1999.
36. Tsai AG, Friesenecker B, Intaglietta M: Capillary flow impairment and functional capillary density. Int J Microcirc Clin Exp 15:238–243, 1995.
Keywords:
Oscillatory blood flow; videomicroscopy; endothelin-1
©2004The Shock Society