INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

VASOMOTION DEFINED AS regular rhythmic oscillations of arteriolar microvessel diameters has been described by many authors (3, 20, 28) using different techniques under physiological and pathophysiological conditions. Vasomotion induces alterations in red blood cell velocity called flow motion, which affects microcirculatory transit times and facilitates gas exchange between microvessels and tissue (14). In previous studies, we (13) observed rhythmic oscillations in microvascular hemoglobin O₂ saturation (HBjO₂) within the jejunal microcirculation in hemodynamically stable pigs. These rhythmic oscillations were unrelated to systemic hemodynamic parameters, respiratory frequency, and intestinal peristalsis and occurred at a frequency of 3.4-5 cycles/min suggesting a flow motion-related phenomenon.

Dopaminergic drugs such as dopamine and fenoldopam significantly attenuated flow motion and at the same time increased jejunal microvascular hemoglobin O₂ saturation and mucosal tissue PO₂ (PO_2muc) suggesting a change in microcirculatory blood flow pattern from alternate periodic vasoconstriction/vasodilation to fixed vasodilation (8).

Recent investigations suggest that flow motion becomes more obvious at the lower range of arterial blood pressure at which blood flow autoregulation is seen. Therefore microvascular hemoglobin O₂ saturation and PO_2muc were recorded in a pig model of stepwise hemorrhage and retransfusion with and without treatment with dopamine to investigate the relationship between the magnitude of flow motion and tissue oxygenation. We hypothesized that flow motion activity is dependent on tissue oxygenation and that dopamine is able to attenuate flow motion during hemorrhage in the pig jejunum. Flow motion within the jejunal wall was indirectly determined by analyzing tracings of HBjO₂, whereas tissue oxygenation was assessed by measurement of PO_2muc using Clark-type multiwire O₂ electrodes.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Animal preparation. Animal experiments were approved by the National Ministry of Science and Research. Thirteen domestic pigs weighing between 35 and 40 kg were fasted for 12 h with free access to water. The animals were anesthetized with 20 mg/kg im ketamine HCl, orally intubated, and mechanically ventilated with a positive end-expiratory pressure of 5 cmH₂O. Tidal volume and respiratory frequency were adjusted to maintain normocapnia; inspiratory O₂ concentration was chosen to keep Pa_O2 levels between 100 and 150 mmHg. Anesthesia was continued with an infusion of 0.8 mg · kg¹ · min¹ midazolam and 20 µg · kg¹ ·min¹ fentanyl. Neuromuscular block was produced by bolus injections of 0.15 mg/kg vecuronium.

Measurement techniques. Arterial pulmonary artery and central venous pressure were measured using pressure transducers (model P10EZ, Spectramed-Statham; Bilthoven, The Netherlands). Cardiac output was determined in triplicate by the thermodilution method. Heart rate, blood pressure, and core temperature were continuously recorded. Arterial, central venous, and mesenteric venous blood gases and acid-base status were analyzed using an automatic blood gas analyzer (model 995, AVL Biomedical Instruments; Graz, Austria). Hemoglobin O₂ saturation was measured with a hemoximeter (Cooximeter, AVL Biomedical Instruments). Hemoglobin concentration and microhematocrit were determined by standard hematological methods.

Measurements of jejunal tissue oxygenation. Methodology for the measurement of PO_2muc and HBjO₂ has been described in detail in previous studies (8, 13). Briefly, PO_2muc was measured by two Clark-type multiwire surface electrodes (Eschweiler; Kiel, Germany), which were calibrated using pure nitrogen and room air in a 37°C warmed water bath. A single electrode consists of eight platinum wires, each 15 µm in diameter, representing eight individual measuring points and an Ag-AgCl reference electrode. An Erlangen microlight guide spectrophotometer (EMPHO II, BGT; Überlingen, Germany) was used for determination of HBjO₂. The measuring principle is based on the use of one illuminating and six detecting microlight guides (each 250 µm in diameter) and a rapidly rotating bandpass interference filter disk for the generation of monochromatic light in 2-nm steps within the spectral range of 502-628 nm representing 64 different wavelengths. Absolute values of HBjO₂ were calculated with the use of an algorithm described in detail by Frank et al. (6), which has been validated in a previous study (13). All sensors are introduced into the measurement chamber and onto the mucosal surface via small openings at the top of the chamber, which are sealed with small caps after the sensors have been correctly positioned. All sensors were kept in place by adhesion using small polyvinylchloride caps, which hold the sensor and are surrounded by a transparent thin (~2 cm in diameter) rubber patch to avoid exposure to room air.

Experimental procedure. After surgical preparation and a resting period of 120 min, baseline measurements of hemodynamics, blood gases, and intestinal tissue O₂ supply were performed [time (t) = 0 min]. Animals were randomized to one of two experimental groups: group C (n = 7) served as controls, whereas group D (n = 7) animals received a continuous intravenous infusion of dopamine (16 µg · kg¹ · min¹) beginning after baseline measurements and continued throughout the experimental period. Animals were given Ringer lactate and gelatin intravenously to maintain PAOP at 12-14 mmHg. At t = 30 min, infusion of Ringer lactate and gelatin was stopped and 45% of the calculated blood volume was removed in three equal steps in all animals at t = 60, 90, and 120 min, respectively. Measurements of systemic and regional parameters were repeated after every bleeding step. After 30 min (t = 150 min), the shed blood was retransfused and an additional infusion of crystalloids and colloids was given to restore pulmonary artery occlusion pressure to baseline values. Measurements were performed at t = 180 min and t = 210 min. At the end of the experiments anesthetized animals were euthanized by central venous bolus injection of 40 mmol KCl.

Statistical analysis. Results are presented as means ± SD. Comparison between baseline values was made using unpaired t-test. Overall effects within and between groups were evaluated by repeated measurements of variance (ANOVA). In case of significant differences, further comparisons were made with paired t-test (within group to baseline) and unpaired t-tests (between groups at individual time points). Values of P  0.05 were considered significant. The Bonferroni-Holm procedure was used for correction of multiple comparisons.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

No differences in systemic hemodynamics, O₂ transport variables, and acid-base status were observed between group C and D animals at baseline (Table 1). Stepwise hemorrhage significantly decreased mean arterial pressure, cardiac index, pulmonary capillary occlusion pressure, and systemic O₂ delivery in both groups, whereas retransfusion restored these parameters to baseline values. Systemic O₂ consumption and arterial blood gas variables remained constant throughout the experiment.

There were no differences in baseline PO_2muc, HBjO₂, mesenteric venous pH, mesenteric venous PO₂ (PO_2mv), and jejunal O₂ extraction ratio between control and dopamine-treated animals (Table 2). Hemorrhage significantly decreased PO_2muc from 33 ± 2.8 mmHg (t = 0 min) to 13 ± 1.6 mmHg (t = 150 min), HBjO₂ from 53 ± 4.9% (t = 0 min) to 32 ± 3.9% (t = 150 min), and PO_2mv from 55 ± 1.8 mmHg (t = 0 min) to 41 ± 1.5 mmHg (t = 150 min) in group C animals. Jejunal O₂ extraction ratio (O₂ER) significantly increased from 0.26 ± 0.02 to 0.51 ± 0.03. During reperfusion, PO_2muc and HBjO₂ remained low in group C animals. Dopamine treatment significantly increased PO_2muc from 28 ± 4.3 mmHg (t = 0 min) to 45 ± 4.6 mmHg (t = 30 min), HBjO₂ from 54 ± 5.7% (t = 0 min) to 69 ± 1.5% (t = 30 min), and PO_2mv from 54 ± 4 mmHg (t = 0 min) to 68 ± 3.9 mmHg (t = 30 min). O₂ER significantly decreased from 0.25 ± 0.05 (t = 0 min) to 0.12 ± 0.03 (t = 30 min). PO_2muc, HBjO₂, and PO_2mv remained significantly higher and O₂ER significantly lower during hemorrhage and retransfusion in group D animals compared with controls. There were no significant differences in PCO_2mv and pH between groups.

Figure 1 demonstrates representative original tracings of HBjO₂ from one group C and group D animal. During hemorrhage and retransfusion, group C animals developed rhythmic oscillations of HBjO₂ with increasing oscillatory amplitude. In contrast, dopamine infusion increased HBjO₂ and prevented development of major HBjO₂ oscillations.

View larger version (24K): [in this window] [in a new window]   Fig. 1.   Original tracings of microvascular hemoglobin O2 saturation (HBjO2) in a control animal (A) and a dopamine-treated animal (B).

Changes in HBjO₂ oscillation amplitude within groups I-III are shown in Fig. 2. Group II oscillations (4.5-7.6 cycles/min) progressively increased with the severity of hemorrhage and remained high even after retransfusion in group C animals, whereas infusion of dopamine prevented the increase in HbjO₂ oscillations.

View larger version (30K): [in this window] [in a new window]   Fig. 2.   A and B: changes in HBjO2 oscillation amplitude within the frequency groups I (1.27-4.4), II (4.5-7.6), and III (7.7-10.5). Values of oscillatory amplitudes are means ± SD. * P  0.05 vs. baseline;  P  0.05 vs. control animals. cpm, Cycles per minute.

The relationship between mean PO_2muc values and amplitudes of HBjO₂ oscillations in the frequency range of 4.5-7.6 cycles/min are shown in Fig. 3. In contrast to group D animals, we observed a significant inverse relationship between PO_2muc and HBjO₂ oscillation amplitude in group C animals.

View larger version (17K): [in this window] [in a new window]   Fig. 3.   Relationship between mucosal tissue PO2 (PO2muc) and amplitude of HBjO2 oscillations in the frequency range of 4.5-7.6 cycles/min in control (A) and dopamine-treated animals (B).

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

This is the first study to demonstrate an inverse relationship between tissue O₂ supply as reflected by jejunal PO_2muc and flow motion activity as within the jejunal wall in a model of stepwise hemorrhage and subsequent fluid resuscitation. Minor oscillations at baseline progressively increased in amplitude without significant changes in frequency, suggesting increased arteriolar vasomotion within the jejunal microcirculation. Despite fluid resuscitation and normalization of systemic hemodynamics and O₂ transport parameters, mean PO_2muc remained low and increased flow motion persisted in control animals. Treatment with continuous infusion of 16 µg · kg¹ · min¹ dopamine significantly increased jejunal tissue oxygenation at baseline and attenuated flow motion during hemorrhage and resuscitation. Dopamine significantly preserved PO_2muc at 15% and 30% blood loss when compared with controls. However, after 45% blood removal, there were no differences in mean PO_2muc between groups.

The biological mechanisms responsible for flow motion are still under investigation. It has been shown that voltage-operated L-type Ca²⁺ channels, cGMP, the Na⁺/K⁺ pump, and the sarcoplasmatic reticulum are part of a biological oscillator system forming the basis for arteriolar vasomotion (1, 10, 11). Electromechanical coupling with oscillations of smooth muscle cell membrane potential, intracellular Ca²⁺, and arteriolar diameter has been demonstrated during hyperoxia in the hamster cheek pouch preparation in vivo (29). Blockade of L-type Ca²⁺ channels with nifedipine immediately attenuated oscillations in membrane potential, intracellular Ca²⁺ and abolished vasomotion, thus pointing at a key role of cell membrane Ca²⁺ channels in the regulation of vasomotion. Recently, Peng et al. (22) hypothesized that arteriolar vasomotion is initiated when intermittent unsynchronized release of Ca²⁺ from the sarcoplasmatic reticulum of vascular smooth muscle cells becomes synchronized in the presence of intact endothelium and under certain conditions.

Unfortunately, mechanisms contributing to the evolution of regular, rhythmic flow motion during periods of limited O₂ supply to tissue are much less clear. Under in vitro conditions, moderate hypoxia has been shown to inhibit L-type Ca²⁺ channels in arterial smooth muscle cells leading to a decrease in intracellular Ca²⁺, promoting muscle relaxation without vasomotion (5). In contrast, in the hamster skeletal microcirculation, moderate to severe hypoxia has been shown to increase the frequency of arteriolar rhythmic diameter changes (2). Increased flow motion was suppressed by phentolamine, a selective -adrenoceptor blocker (4).

Because of contrasting experimental results, we can only speculate on the mechanisms of increased flow motion during ischemia and reperfusion. Despite the lack of final evidence, the association between decreased PO_2muc and increased oscillations in HbjO₂ in face of normalized systemic hemodynamics and systemic O₂ transport after resuscitation suggests that the magnitude of flow motion is not simply dependent on presence of systemic hypotension or low systemic blood flow. One might speculate about the existence of a cause effect relationship between low tissue PO₂ and flow motion activity involving an O₂-sensing system within the jejunal microcirculation.

Dopamine was able to block flow motion even in the presence of low tissue PO₂. Experimental studies (7, 9, 18) have shown that catecholamines interfere with arteriolar vasomotion. Norepinephrine, for example, induces vasomotion by converting unsynchronized intracellular Ca²⁺ oscillations into global synchronized changes in vascular smooth muscle cells in the presence of endothelium via a cGMP-dependent mechanism (22). In previous experiments, we have demonstrated that the dopaminergic drugs dopamine and fenoldopam increase PO_2muc in a dose-related manner and offset flow motion within the jejunal microcirculation. These effects were blocked by the application of SCH-23390, a selective dopamine-1 receptor antagonist, suggesting that attenuation of flow motion was mediated by a cAMP-dependent mechanism (author's unpublished observation). Therefore it is conceivable that drugs that increase intracellular cAMP within vascular smooth muscle cells interfere with the activity of flow motion in the jejunal microcirculation even in the presence of limited O₂ supply.

Arteriolar vasomotion of microvessels and increased flow motion has been increasingly observed under conditions of low flow, hypotension, and tissue hypoxia. Increased flow motion is believed to provide adequate temporal and spatial tissue perfusion to maintain nutritional blood flow in tissue under limited O₂ supply. Mathematical model analysis demonstrated that hydraulic resistance of blood vessels exhibiting vasomotion is less compared with vessels showing static diameters with identical average (26). Furthermore, it has been suggested that vasomotion induced flow motion facilitates fluid reabsorption within capillaries and venules during periods of arteriolar vasoconstriction (15). Augmented fluid absorption might be of major importance for attenuating development of tissue edema in particular after ischemia-reperfusion injury. Formation of tissue edema may continuously deteriorate tissue O₂ supply. Rücker et al. (23) demonstrated in a muscle periostium skin preparation in rats that flow motion in critically perfused tissue not only preserves functional capillary density but also protects the adjacent tissue from capillary perfusion failure. Flow motion in different gut segments has been repeatedly demonstrated under physiological and pathophysiological conditions. By means of laser-Doppler velocimetry and reflectance spectrophotometry, Yamaguchi et al. (30) demonstrated regular changes in microcirculatory blood flow, indexes of microvascular hemoglobin oxygenation and concentration at a frequency ranging between 4 and 6 cycles/min in the rat gastric mucosa. These oscillations increased in amplitude and frequency during hemorrhage, whereas gastric motility remained constant. During normothermic cardiopulmonary bypass (CPB), we reported the onset of HBjO₂ oscillations with frequencies of 5-7 cycles/min simultaneously with a significant drop in PO_2muc after institution of CPB (12). There were no significant differences in mean arterial pressure and systemic O₂ delivery between sham and CPB animals. Therefore, increased flow motion most likely resulted from local alterations in jejunal blood flow induced by the onset of CBP.

Methods for assessing tissue oxygenation and flow motion activity. Measurements of organ surface PO₂ using Clark-type electrodes have been established in several studies as a suitable instrument for assessing tissue oxygenation in a variety of organs (7, 17, 19, 21). PO_2muc measurements require diffusion of O₂ through mucosal epithelial cells to the electrode were O₂ is consumed in a complex redox reaction (16). The PO₂ measured reflects the PO₂ in underlying cells, as long as diffusion error of the electrode is small and the pollution of electrode surface by atmospheric O₂ can be excluded. O₂ consumption of multiwire surface electrodes is small. Measurement errors due to diffusion of atmospheric O₂ to the electrode surface can be avoided with the measurement setup used (13). Therefore, multiwire surface electrodes reflect tissue PO₂ present in mucosal epithelial cells underneath the electrode. On the basis of scanning electronic microscopy investigation results of microvasculature of jejunal villi in pigs and taking into account the surface area of 19 mm² of a multiwire surface O₂ electrode, the electrode roughly covers a mucosal area containing 380 villi (13).