INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

THE PRESSURE-FLOW (P-) curve has been widely used in many studies to describe the characteristics of blood vessels and hemodynamics. Various preparations have been designed to measure the relationship between pressure and flow in the pulmonary circulation, including 1) isolated lung (10) or lobes (1, 2), 2) in situ perfused lung with either a systemic arteriovenous fistula (4) or extracorporeal pump (12), 3) intravascular balloon occlusion (6, 8), or 4) combination of arteriovenous fistulas and occlusion balloon in the inferior vena cava to vary cardiac output (15, 17). In the isolated lung or lobes, it is easy to control the blood flow and to continuously monitor the blood pressure (19). Because the isolated lung is stable only for several (~3-4) hours after its isolation, studies are limited to acute injury. In addition, several reports in the literature indicate that an isolated lung may predispose to an increase in membrane permeability (20, 22, 24), although other investigations have obtained contrary results (5).

Intravascular balloon inflation usually has been used in middle-sized (14) or large animal models (6, 15, 17). There is no report on the use of this balloon technique in the left pulmonary artery in rats. Apart from these preparations mentioned above, blocking unilateral blood flow and then reperfusing one side of the lungs in intact animals has elicited much discussion and interest (7, 21, 23). Under these circumstances, it is more meaningful to measure the P- curve of a single lung. Recently, Hyman et al. (12) used a balloon to partially block blood flow of the right lung and then used an in vitro pump to control different flows from the right atrium or the aorta to reperfuse the partial right lung. However, Hyman et al. (12) failed to induce a reperfusion injury in their study. Therefore, in this current in vivo study, we tested whether ischemia-reperfusion (IR) induces unilateral vascular changes in rat lungs using the P- curve.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Animal preparation. Twenty-four male Wistar rats weighing 427 ± 7 g were randomly separated into three groups: sham (n = 8), IR1 (n = 9), and IR2 (n = 7). Rats in both IR groups were subjected to the IR procedure (see below for detail) in the left lung. Animals in the IR1 and IR2 groups were used to perform the P- curve (see below for detail) in the left and the right lung, respectively. Rats in the sham group received the same operations as the IR group without IR procedure. The rats were anesthetized with pentobarbital sodium (35 mg/kg, ip). A Silastic catheter (Silastic medical-grade tubing, 0.305 mm ID, 0.635 mm OD) was inserted via the internal jugular vein, and the tip was placed near the right atrium for injection of 15-µm colored microspheres (E-Z TRAC) to measure blood flow in each lung. Tracheal cannulation was performed after tracheostomy. The chest was opened via a midline incision, and the animal was ventilated by a ventilator with room air, at a frequency of 70/min and tidal volume of 10 ml/kg, with ~2-3 cmH₂O positive end-expiratory pressure (PEEP). The bronchus and blood vessels supplying the left lung and those supplying the two middle parts of the right lung were wrapped with a loop of PE-10 tube linked to another tube (5-6 mm ID). During the ischemia, the loop was tightly pulled to simultaneously stop blood flow and ventilation. A PE-10 tube was directly punctured into the right ventricle outflow tract and another one into the left atrium. Subsequently, Silastic catheters were separately inserted, via the above puncture marks, for monitoring of the left pulmonary artery and left atrium blood pressures, respectively. However, for simplifying the surgery, we omitted the process for monitoring the blood pressure of left atrium in the IR2 group. To detect precise changes of blood pressure, we pushed the Silastic tube 0.5 cm inside the left or right pulmonary artery to measure both the pulmonary pressure and the pressure at zero flow toward the left or right lung (wedge pressure). The wedge pressure was utilized to indicate the change in closing pressure of the vascular bed. Zero flow was obtained by lifting a thread wrapped around the left pulmonary artery. Both wedge pressure and atrial pressure were detected with a pressure transducer (DTX/Plus, Viggo-Spectramed), which was connected to a MacLab 200 data recorder (ADInstruments).

The animal model of lung ischemia-reperfusion injury. Basically, each IR study consisted of three sequential periods: baseline, ischemia, and reperfusion. The intervals of baseline, ischemia, and reperfusion periods were 30, 90, and 30 min, respectively. At the end of the baseline period, each animal received intravenously 50 units of heparin (total volume of 500 µl) in saline via the jugular vein. Subsequently, the ventilator was removed briefly (10-20 s), and the ischemia lung (left lung) was slightly collapsed by pressing with wet cotton (21) before vascular occlusion with the loop apparatus mentioned above. The counterlateral lung was continuously ventilated. At the beginning of the reperfusion period, the loop apparatus was removed and the animal was ventilated with pure oxygen. Thirty minutes after the start of reperfusion, the left atrial appendage was cut and the lungs were flushed with 50 ml of saline via the right ventricle using a gravity pressure of 30 cmH₂O. The lungs were then removed to analyze the wet weight-to-dry weight ratio and microsphere number (see Comparison of pulmonary blood flow measured by the microsphere and the ultrasonic methods and Wet-to-dry weight ratio analysis).

P- curves. P- curves were obtained during both the baseline and reperfusion periods, and each curve was constructed by plotting three different blood flows against simultaneously obtained pressure differences between the left pulmonary artery and left atrium pressures. Because it was difficult to cannulate the pulsating left atrium, left atrial pressure monitoring was omitted for P- curves of the right lung. Four different types of colored microspheres were randomly used in this study. Each type, containing 5 × 10⁴ microspheres, was suspended in 5 µl saline with 0.05% vol/vol Tween 80 and 0.01% wt/vol thimerosal. The first value of blood flow was obtained by administering one type of colored microspheres from the jugular vein catheter without any vascular occlusion. The second value (elevated blood flow in the left lung) was obtained by administering another type of colored microspheres during occluded blood supply to the two middle lobes of the right lung. The different blood flows of the left and the right lungs were calculated as explained in detail in the next section, according to the number of colored microspheres administered (5 × 10⁴ microspheres). The cardiac output was determined by a Transonic Flowmeter (T-106, Transonic System). The final value of zero blood flow was obtained without using colored microspheres by suspending the left or right pulmonary artery to stop its blood flow. The same processes were also repeated during the reperfusion period.

Comparison of pulmonary blood flow measured by the microsphere and the ultrasonic methods. In total, five rats weighing 428 ± 7 g were used to compare the pulmonary blood flow measured by the microsphere and the ultrasonic methods. After tracheal and jugular vein cannulation, the chest was opened and the anesthetized animal was ventilated as described above. Two ultrasonic probes (2SB and 1RB series) were separately mounted on the ascending aorta and left pulmonary artery to measure blood flow as read directly from a Transonic flowmeter (T-206, Transonic Systems) with a self-built digital display. The right lung blood flow was calculated by subtracting the left lung blood flow from the total blood flow (cardiac output) measured in the ascending aorta. For the microsphere method, all colored microsphere preparations were similar to that described under P- curves. Over 2 h, microspheres were administered into the jugular vein four times (each time with only one type of colored microsphere), with any two injections separated by 30 min. After each injection, 25 µl saline (containing 2% vol/vol Tween 80) was used to flush the dead space of the tube. At the end of the experiment, an overdose of pentobarbital was administered to kill the animal. The left and right lungs were isolated for microsphere extraction, described in Extraction of colored microspheres from lung tissue. By using measured microsphere numbers, the percentage of blood flow perfusing either the left or right lung was calculated and then correlated with that of the percentage of blood flow measured with the ultrasonic method.

Wet-to-dry weight ratio analysis. When the lungs were harvested, the left lung was separated from the right lung, and the wet weight of each lung was obtained. The lung was then dried to a constant dry weight in an oven kept at 50°C. The wet-to-dry weight ratio was then determined.

Extraction of colored microspheres from lung tissue. Dried lung was dissolved by a 4N KOH solution (containing 2% polyoxyethylene sorbitan monooleate, Tween 80) in 90°C water bath for 30 min according to the method of Hakkinen et al. (9) with some modifications. Proper HCl solution (containing 2% Tween 80) was used to neutralize the 4N KOH solution. The microsphere pellet was washed (wash solution: H₂O + 2% Tween 80) two times and counted by using a hemocytometer.

Data analysis. All data are presented as means ± SE. Simple linear regression was used to establish the correlation between blood flows measured by the microsphere and ultrasonic methods (3). Slopes of the P- curves were fitted first with simple linear regression and then compared by paired or unpaired Student's t-test.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Effects of IR. Figure 1 depicts the rotation of P- curves caused by IR in the IR1 group. After IR, these curves significantly rotated to the left. In contrast, IR caused no significant change in the intercept on the vertical axis. Mean r values of linear equations for P- curves (Fig. 1) in the baseline and reperfusion periods were 0.99 ± 0.03 and 0.95 ± 0.04, respectively. P- curves before and after a null-ischemia in the sham group are shown in Fig. 2. No marked rotations in the P- curves were found after the null-ischemia. Mean r values of the linear equations for P- curves before and after the null-ischemia were 0.89 ± 0.03 and 0.93 ± 0.03, respectively. Figure 3 illustrates P- curves of the right lung in the IR2 group. These curves were similar to those of the left lung in the sham group, with no marked changes in these curves after the left lung ischemia. Figure 4 combines all slopes shown in Figs. 1 and 2. Compared with the baseline period of both IR1 and sham groups, there was a significant increase in the slope of the P- curve in the ischemic left lung of the IR1 group, implying IR-induced vasoconstriction. In addition, using the P- curves shown in Figs. 1 and 2, we calculated vascular resistance at two blood flow values, 5 and 15 ml/min (Fig. 5). IR caused significant increase in vascular resistance at both of these two flow rates (Fig. 5).

View larger version (27K): [in this window] [in a new window]   Fig. 1.   Pressure-flow curves [(Ppa  Pla) vs. left pulmonary arterial blood flow] of the left lung before (baseline, dashed lines) and after (solid lines) ischemia-reperfusion (IR). Ppa, left pulmonary arterial pressure; Pla, left atrial pressure.

View larger version (28K): [in this window] [in a new window]   Fig. 2.   Pressure-flow curves [(Ppa - Pla) vs. left pulmonary arterial blood flow] of the left lung before (baseline, dashed lines) and after (solid lines) null-ischemia reperfusion in the sham group.

View larger version (31K): [in this window] [in a new window]   Fig. 3.   Pressure-flow curves of the right lung before (baseline, dashed lines) and after (solid lines) left lung ischemia reperfusion in the IR2 group.

View larger version (25K): [in this window] [in a new window]   Fig. 4.   Comparison of slopes of pressure-flow curves of the left lung between the ischemia-reperfusion (IR1) group and the sham group.

View larger version (19K): [in this window] [in a new window]   Fig. 5.   Left vascular resistances at 2 flow rates in 2 groups of rats. Statistical difference (P < 0.01) compared with the respective baseline value.

Comparison between methods. There was a significant correlation between the percentage of blood flow measured by the microsphere and by the ultrasonic methods in the left lung (Fig. 6). A similar significant correlation was also observed in the right lung (Fig. 7). Figure 8 shows the significant correlation between the two flow measurements using the microsphere and ultrasonic methods in both right and left lungs.

View larger version (25K): [in this window] [in a new window]   Fig. 6.   Linear regression of left pulmonary blood flow (expressed as percentage of total blood flow) measured simultaneously by the microsphere and the ultrasonic methods in 5 rats.

View larger version (24K): [in this window] [in a new window]   Fig. 7.   Linear regression of right pulmonary blood flow (expressed as percentage of total blood flow) measured simultaneously by the microsphere and the ultrasonic methods in 5 rats.

View larger version (20K): [in this window] [in a new window]   Fig. 8.   Linear regression plot of both right and left blood flows measured simultaneously by the microsphere and ultrasonic methods in 5 rats.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

The main purpose of this study was to use the retention of the colored microspheres in the left and right lungs to detect the blood flow in each lung before and after unilateral (left) IR. From the measured and calculated blood pressures and blood flows of each lung, we plotted P- curves. By using the P- curve, we described IR-induced changes in pulmonary vascular characteristics.

The reason that we cannulated the left pulmonary artery instead of the main pulmonary artery was because the left pulmonary arterial pressure may be different from the main pulmonary arterial pressure (27). Advantages of choosing P- curves for the evaluation of vascular changes in this study include the simplicity of this technique, its ease of performance, and the fact that it does not need complicated equipment or operations. One disadvantage of using P- curves may be that the data values are not sufficient to show the whole curve, such as the one reported by Hyman et al. (12). In our study, the averaged blood flow value of the left lung for both the ischemia and the sham groups during the baseline period was 10.64 ± 0.76 ml/min. After partial blocking the flow of the right lung, blood flow in the left lung increased to 15.17 ± 1.47 ml/min. In terms of blood flow per unit of lung volume, our range of observed blood flow in the left lung locates on the lower half of the P- curve of Hyman et al. (12). This lower half portion of the curve is relatively rectilinear.

The microsphere and ultrasonic methods are commonly used to detect local blood flow. For the microsphere method, we used the same-sized microspheres (15.00 ± 0.31 µm). We injected ~5 × 10⁴ microspheres and recovered 4.1 × 10⁴ ± 1,203 microspheres (mean ± SE, n = 56) from the lungs. We did a simple experiment to check microspheres that were not recovered from the lungs. Two additional rats were injected with four different-colored microspheres at four different times. Subsequently, brain, liver, and kidney were examined for microspheres. After the injection of 5 × 10⁴ microspheres, the average numbers of microspheres retained in the injection system (syringe and tubing) were 7,000 ± 707 (14.0%). No microspheres were found in either brain, liver, or kidney, however. At high flow, such as flow >10 ml/min, variation in flow became larger, and thus correlation between results of the microsphere method and ultrasonic method diverges widely (Fig. 8). Similar large variation at high flow was also noted by Kowallik et al. (13). For the ultrasonic method in this study, a probe (2SB or 1RB series) of the ultrasonic flowmeter has to be cuffed on a totally isolated vessel. The right pulmonary artery is hard to isolate because it is directly under the aorta. Although the length of the left pulmonary artery is barely long enough for the probe to be cuffed, the length of the left pulmonary artery in each individual animal was different. Furthermore, the cuffed probe could be squeezed by lung expansion during inflation, resulting in alteration of the angle between the probe and pulmonary artery. The above unusual conditions, although they were rare, might predispose the ultrasonic flowmeter to detect blood flow inaccurately and cause the correlation coefficient between blood flows measured by the two methods to be relatively low (Fig. 6). The plotting between the two flow measurements using the microsphere and ultrasonic methods has a high correlation (Fig. 8). This may indicate that the microsphere method is also a good way to detect pulmonary blood flow.

It has been verified by several previous studies that IR can cause an increase in resistance of pulmonary vessels (16, 26). To our knowledge, this is the first study to demonstrate, in vivo, IR-induced unilateral vascular changes in rat lungs using the P- curve. Our experimental results indicate that IR caused a leftward rotation of the P- curve (around a fixed y-axis). However, the null-IR did not induce the same rotation in the sham group. Mitzner and Chang (18) elegantly analyzed many factors (such as rigid or distensible vessels, Starling resistors, arteriovenous shunt, etc.) on the P- curve in detail. For our purpose, two ways were used to analyze the P- curve: the slope of the curve and the intercept on the vertical axis. The slope of the curve represents the pressure gradient needed to move a fixed blood flow. When the curve rotates to the left, it indicates an increase in the pressure difference required to move a fixed amount of blood flow due to vascular constriction. This may also be accompanied by the same pressure difference with a decreased blood flow. The alteration of interception may show the change of perivascular pressure that collapses the pulmonary resistant vessels. The difference between pulmonary arterial pressure (Ppa) and left atrial pressure (Pla) is commonly used as a vertical axis on a P- curve. In this study, the difference between Ppa and Pla at zero flow was near zero and did not change after IR. At zero flow, the similar values of Ppa and Pla might reflect that pulmonary vessels are not completely compressed (18).

In this study, IR induced a significant increase in Ppa. Also, IR caused a significant increase in lung wet weight-to-dry weight ratio, indicating edema formation in the left lung. When we examined these data for the entire study, resistances of edematous left lungs were usually higher than those of sham-treated lungs. However, we did not find a significant correlation between the degree of edema and vascular resistance in the left lung. Similarly, Wang et al. (25) demonstrated no correlation between edema and vascular resistance in partially isolated dog lungs. Therefore, our results and those of Wang et al. (25) are consistent with the suggestion made by Hogg (11) that fluid in the bronchovascular interstitial space has little effect on blood flow and that gross alveolar edema is required before an appreciable increase in vascular resistance. In this context, it was reasonable to learn that severe edema facilitated an increase in pulmonary vascular resistance in the in vitro studies (2, 4).

In summary, we found IR-induced changes in pulmonary vascular characteristics by using the P- curve. In addition, we demonstrated a significant correlation between blood flow values obtained by the microsphere and ultrasonic methods in both right and left lungs.