Halothane Impairs the Hemodynamic Influence of Endothelium-derived Nitric Oxide 

Male Sprague-Dawley rats (Charles River Laboratory, Wilmington, MA) weighing 225–325 g were used in all experiments. Rats fasted overnight but were allowed free access to water. Cardiac output and regional blood flow were determined using radiolabeled microspheres.

In all experiments, a PE 10 catheter (Fisher Scientific, Chicago, IL) was passed through the right common carotid artery into the left ventricle for infusion of microspheres. The catheter tip was adjusted until pulse pressure in the left ventricle could be read without artifacts. In addition, the right femoral vein and artery were catheterized with PE 50 tubing; the venous catheter was used for constant infusion of saline (40 micro liter/min) in anesthetized rats and for drug infusion and blood replacement in all rats. The arterial catheter was used to monitor BP and for reference blood sampling. Blood pressure and heart rate (HR) were monitored using a Statham pressure transducer (Viggo-Spectramed, Oxnard, CA) connected to a Gould chart recorder (Gould, Valley View, OH). Heart rate in beats per minute (beats/min) was derived electronically from the arterial pressure wave.

The effect of EDNO synthesis inhibition on systemic and regional hemodynamics was measured using paired injections of radioactive microspheres (DuPont-New England Nuclear, Boston, MA), 15 plus/minus 1.5 micro meter in diameter, labeled with either Cerium¹⁴¹or Strontium⁸⁵, as described previously. [20] Because of rats' severe hypotensive reaction to the commercial dextran vehicle, [21] microspheres were suspended in 3.5 M glucose using 0.01% Tween-80 as an antiaggregant. We have found that this concentration has no effect on systemic pressure. Microspheres at a concentration of 400,000/ml were agitated mechanically for approximately 15 min. Then 0.2 ml of the suspension, corresponding to about 80,000 microspheres, was withdrawn into a syringe that was counted to obtain the preinjection dose. The microspheres, together with 0.2 ml saline, were infused into the left ventricle over 20 s; at the same time, arterial blood was withdrawn at a rate of 0.48 ml/min over 75 s. The sampled blood was replaced with blood from a donor rat nephrectomized 16–24 h earlier. After microsphere injection, the syringe was again counted to obtain the postinjection dose. The injection count was obtained by subtracting the preinjection from the postinjection dose.

The first injection of microspheres was given under stable control conditions. Fifteen minutes after EDNO synthesis inhibition, the second set of microspheres was injected. At the conclusion of the experiment, the animals were killed with an overdose of pentobarbital sodium (Nembutal, Abbott Laboratories, Chicago, IL; 150 mg/kg intravenously) and the target tissues excised for analysis. Samples were counted in a Packard gamma counter using dual window settings of 10–250 and 400–700 MeV at a sample level of 0.5 cm.

A bolus dose of 10 mg/kg body weight (bw) of L-NAME (Sigma, St Louis, MO) was used to inhibit EDNO synthesis. We have previously shown that this dose induces sustained EDNO synthesis inhibition in the systemic and renal vasculature. [22] We used the response to L-NAME as an index of the degree of participation of endogenous EDNO in regulating systemic and regional hemodynamics. All experimental protocols were approved by the Institutional Animal Care and Use Committee and are consistent with the United States Public Health Services Guidelines on Care and Use of Laboratory Animals.

In conscious rats, the catheters were implanted the day before as described above, using diethyl ether anesthesia and antiseptic technique. The catheters were led subcutaneously and exteriorized at the nape of the neck. After fasting overnight, rats were isolated in restraining cages for 60 min, during which BP and HR were monitored. A separate group of rats were instrumented and blood pH, p^CO2, and p^O2were measured before and after L-NAME.

In the barbiturate-anesthesia group, rats fasted overnight were anesthetized by an intraperitoneal injection of 125 mg/kg bw thiobutabarbital (Inactin, Andrew Lockwood Co., Milwaukee, WI) and placed on a heating pad to maintain constant body temperature. Rats were instrumented as described above and allowed a 60-min stabilization period, during which BP and HR were monitored. A separate group of rats were instrumented and blood pH, p^CO2, and p^O2were measured before and after L-NAME. All rats spontaneously breathed room air without the aid of mechanical ventilation.

In halothane-anesthetized rats, the catheters were implanted the day before as described above and exteriorized at the nape of the neck. After fasting overnight, halothane was administered in an air-tight, 5–1 Plexiglas chamber using a calibrated vaporizer which delivered 1% halothane with 21% Oxygen²at a gas flow of 4 l. After anesthesia, rats were allowed a 60-min stabilization period, during which BP and HR were monitored. A separate group of rats were instrumented and blood pH, p^CO2and p^O2were measured before and after L-NAME. All experiments were performed on spontaneously breathing rats.

Cardiac output (CO; ml *symbol* min sup -1 *symbol* 100 sup -1 *symbol* g bw), organ blood flow (ml *symbol* min sup -1 *symbol* g tissue weight sup -1), total peripheral resistance (TPR; mmHg *symbol* ml sup -1 *symbol* min sup -1 *symbol* 100 sup -1 g bw, referred to as resistance units or RU), and organ vascular resistance (mmHg *symbol* ml sup -1 *symbol* min sup -1 *symbol* g tissue weight sup -1, referred to as resistance units or RU) were determined from the measured counts per minute (CPM) as follows: 1) CO = cpm injected x pump speed *symbol* cpm blood per 100 g bw sup -1, 2) organ blood flow = cpm organ x pump speed/(cpm blood x tissue weight [g]) sup -1, 3) TPR = mean BP/CO, and 4) organ vascular resistance = mean BP/organ blood flow. The right and left kidneys were analyzed separately to establish a control for the technique. To accept the results of the experiment, blood flow in each kidney had to be within 10% of the mean calculated value of both kidneys.

Data were evaluated in several statistical ways. To assess the influence of L-NAME, Student's paired t test was run on the changes resulting from treatment; P < 0.05 was considered significant. Comparisons among the three groups (conscious and barbiturate- and halothane-anesthetized) were analyzed using ANOVA and the Bonferroni's t test; an adjusted P value of P < 0.05 was considered significant.

Changes in BP, HR, CO, and TPR in conscious rats in response to L-NAME are shown in Figure 1and Figure 2. N^W-nitro-L-arginine methyl ester caused a 24 plus/minus 2 mmHg increase in BP ( P < 0.001) and a 146 plus/minus 12-beats/min decrease in HR (P < 0.001). Cardiac output decreased 47%(P < 0.005), while TPR increased 132%(P < 0.001).

Changes in regional hemodynamics in conscious rats in response to EDNO synthesis inhibition are shown in Figure 3.

Basal cerebral blood flow was 1.2 plus/minus 0.10 ml *symbol* min sup -1 *symbol* g tissue weight sup -1 and vascular resistance was 96 plus/minus 9 RU. N^W-nitro-L-arginine methyl ester decreased blood flow by 23%(P < 0.005) and increased vascular resistance by 67%(P < 0.02).

Basal cardiac blood flow was 5.7 plus/minus 0.6 ml *symbol* min sup -1 *symbol* g tissue weight¹, while vascular resistance was 22 plus/minus 3 RU. N^W-nitro-L-arginine methyl ester decreased blood flow by 30%(P < 0.02) and increased vascular resistance by 95%(P < 0.001).

Basal hepatic arterial blood flow was 0.4 plus/minus 0.1 ml *symbol* min sup -1 *symbol* g tissue weight sup -1, while vascular resistance was 391 plus/minus 59 RU. N^W-nitro-L-arginine methyl ester had no significant effect on either blood flow or vascular resistance.

Basal renal blood flow was 8.4 plus/minus 0.9 ml *symbol* min sup -1 *symbol* g tissue weight sup -1, while vascular resistance was 15 plus/minus 2 RU. N^W-nitro-L-arginine methyl ester decreased blood flow by 47%(P < 0.001) and increased vascular resistance by 145%(P < 0.001).

Basal mesenteric blood flow was 2 plus/minus 0.2 ml *symbol* min sup -1 *symbol* g tissue weight sup -1, while vascular resistance was 60 plus/minus 6 RU. N^W-nitro-L-arginine methyl ester decreased blood flow by 53%(P < 0.001) and increased vascular resistance by 167%(P < 0.001).

Blood pH, p^CO2, and p^O2in conscious rats (n = 3) before and after L-NAME are shown in Figure 4. Under basal conditions, conscious rats had a blood pH of 7.46 plus/minus 0.01, a p sub CO²of 42 plus/minus 2.6 mmHg, and a p^O2of 87 plus/minus 6 mmHg. Administration of L-NAME had no effect on blood pH or p^CO2levels, while causing a slight but significant 6%(P < 0.05) increase in p^O2levels.

Changes in BP, HR, CO, and TPR in inactin anesthetized rats in response to L-NAME are shown in Figure 1and Figure 2. N^W-nitro-L-arginine methyl ester caused a 20 plus/minus 1-mmHg increase in BP (P < 0.001) and a 36 plus/minus 6-beats/min decrease in HR (P < 0.001). Cardiac output decreased by 38%(P < 0.005), while TPR increased by 106%(P < 0.001). The decrease in HR was significantly less than in conscious rats (P < 0.05).

Changes in regional hemodynamics in inactin-anesthetized rats in response to EDNO synthesis inhibition are shown in Figure 3.

Basal cerebral blood flow was 0.6 plus/minus 0.1 ml *symbol* min sup -1 *symbol* g tissue weight sup -1 and vascular resistance was 179 plus/minus 20 RU. N^W-nitro-L-arginine methyl ester had no effect on blood flow, but increased vascular resistance by 44%(P < 0.05). The decrease in blood flow was significantly less than that seen after L-NAME in conscious rats (P < 0.05).

Basal cardiac blood flow was 4.2 plus/minus 0.4 ml *symbol* min¹*symbol* g tissue weight sup -1, while vascular resistance was 25 plus/minus 3 RU. N^W-nitro-L-arginine methyl ester had no effect on either blood flow or vascular resistance, although there was a nonsignificant trend for L-NAME to increase resistance.

Basal hepatic arterial blood flow was 0.4 plus/minus 0.05 ml *symbol* min¹*symbol* g tissue weight sup -1, while vascular resistance was 281 plus/minus 42 RU. N^W-nitro-L-arginine methyl ester decreased blood flow by 48%(P < 0.005) and increased vascular resistance by 146%(P < 0.001). Although the decrease in blood flow after L-NAME was only about one-half that observed in conscious rats, the variability was such that this difference was not statistically significant.

Basal renal blood flow was 5.6 plus/minus 0.4 ml *symbol* min sup -1 *symbol* g tissue weight sup -1, while vascular resistance was 18 plus/minus 1 RU. N^W-nitro-L-arginine methyl ester decreased blood flow by 35%(P < 0.001) and increased vascular resistance by 93%(P < 0.001). The decrease in blood flow was significantly less than that observed in conscious rats (P < 0.05).

Basal mesenteric blood flow was 1.8 plus/minus 0.2 ml *symbol* min sup -1 *symbol* g tissue weight sup -1, while vascular resistance was 59 plus/minus 5 RU. N^W-nitro-L-arginine methyl ester decreased blood flow by 51%(P < 0.001) and increased vascular resistance by 166%(P < 0.005).

Blood pH, p^CO2, and p^O2in inactin-anesthetized rats (n = 3) before and after L-NAME are shown in Figure 4. Under basal conditions, rats had a blood pH of 7.35 plus/minus 0.02, a p^CO2of 59 plus/minus 3.3 mmHg, and a p^O2of 66 plus/minus 1.9 mmHg. Administration of L-NAME had no effect on blood pH, p^CO2, or p^O2levels.

Changes in BP, HR, CO, and TPR in halothane-anesthetized rats in response to L-NAME are shown in Figure 1and Figure 2. N^W-nitro-L-arginine methyl ester caused only a 7 plus/minus 1-mmHg increase in BP (P < 0.001) and no change in HR. Cardiac output did not decrease, while TPR increased by only 22%(P < 0.02). All systemic hemodynamic responses to L-NAME were significantly attenuated compared with either conscious or barbiturate-anesthetized rats.

Changes in regional hemodynamics in halothane-anesthetized rats in response to EDNO synthesis inhibition are shown in Figure 3.

Basal cerebral blood flow was 1.7 plus/minus 0.1 ml *symbol* min sup -1 *symbol* g tissue weight sup -1 and vascular resistance was 53 plus/minus 4 RU. N^W-nitro-L-arginine methyl ester had no effect on either cerebral blood flow or vascular resistance. The blunted cerebral hemodynamic responses were significantly different from the changes observed in either conscious or barbiturate-anesthetized rats (P < 0.05).

Basal cardiac blood flow was 3.8 plus/minus 0.2 ml *symbol* min sup -1 *symbol* g tissue weight sup -1, while resistance was 24 plus/minus 1. N^W-nitro-L-arginine methyl ester had no effect on either blood flow or vascular resistance, although there was a general tendency for L-NAME to increase cardiac resistance. These diminished responses were significantly less than the changes in cardiac hemodynamics observed in the conscious rats (P < 0.05), but not different from the response in barbiturate-anesthetized rats.

Basal hepatic arterial blood flow was 0.3 plus/minus 0.01 ml *symbol* min sup -1 *symbol* g tissue weight sup -1, while vascular resistance was 304 plus/minus 14 RU. N^W-nitro-L-arginine methyl ester had no effect on either blood flow or vascular resistance.

Basal renal blood flow was 4.2 plus/minus 0.2 ml *symbol* min sup -1 *symbol* g tissue weight sup -1, while vascular resistance was 22 plus/minus 1 RU. N^W-nitro-L-arginine methyl ester had no effect on renal blood flow and increased vascular resistance by only 19%, (P < 0.02). The lack of change in blood flow and the increase in vascular resistance were significantly different from the changes in renal hemodynamics observed in either conscious or barbiturate-anesthetized rats (P < 0.05).

Basal mesenteric blood flow was 0.8 plus/minus 0.1 ml *symbol* min¹*symbol* g tissue weight sup -1, while vascular resistance was 107 plus/minus 6 RU. N^W-nitro-L-arginine methyl ester decreased blood flow by 20%(P < 0.01) and increased vascular resistance by 40%(P < 0.01). The decrease in blood flow was significantly less than that seen in either conscious or barbiturate-anesthetized rats (P < 0.05).

Blood pH, p^CO2, and p^O2in halothane-anesthetized rats (n = 3) before and after L-NAME are shown in Figure 4. Under basal conditions, rats had a blood pH of 7.37 plus/minus 0.01, a p^CO2of 48 plus/minus 0.8 mmHg, and a p^O2of 64 plus/minus 1 mmHg. Administration of L-NAME had no effect on either blood pH, p^CO2, or p^O2levels.

The tonic vasodilator EDNO is an important factor in the regulation of vascular tone and organ perfusion. [1,2] In vivo and in vitro data have indicated that some volatile anesthetics may inhibit the vasodilator influence of EDNO, thereby increasing vascular tone. [14,18,19] Our study was designed to test the hypothesis that halothane inhibits EDNO-mediated vasodilation and its regulation of regional blood flow. To test this hypothesis, we inhibited EDNO synthesis in halothane-anesthetized rats and compared systemic and regional hemodynamic responses with conscious and inactin (barbiturate)-anesthetized rats. Regional hemodynamics were measured before and after inhibiting EDNO synthesis with L-NAME. Our results demonstrate that, although EDNO synthesis inhibition increased BP, TPR, and local vascular resistance and decreased CO and HR in both conscious and inactin-anesthetized rats, halothane anesthesia greatly blunted these responses. The observation that blood pH was similar in all three groups and that blood p^O2was not different between inactin- and halothane-anesthetized rats indicates that the highly attenuated systemic and regional hemodynamics responses to EDNO synthesis inhibition in halothane-anesthetized rats was not the influence of a lack of mechanical ventilation and a lowered p^O2, nor just the suppressive influence of anesthesia, but, rather, apparently the result of some halothane-mediated effect. Taken together, these findings support the hypothesis that halothane inhibits the influence of EDNO on systemic and regional vascular tone.

The endothelium modulates vascular tone through the production of various vasoactive factors. [8] Previous reports have demonstrated the importance of basal EDNO production on regional hemodynamics. [9–12] In conscious rats, inhibition of EDNO synthesis results in increased BP and TPR and decreased CO. [12] The rise in TPR occurs concurrently with increased resistance in many vascular beds, including the renal, mesenteric, cerebral, gastric, and hepatic circulations. [12,23] However, these studies indicate that the apparent influence of EDNO in maintaining organ vascular resistance is not uniform among all vascular beds. Responses to L-NMMA range from a 15% decrease in bronchial vascular resistance to a 150% increase in hepatic resistance. [12] Our observations confirm the heterogeneous response of different resistance beds to EDNO synthesis inhibition. In conscious rats, vascular resistance tended to increase in all vascular beds investigated with changes in the visceral beds; in particular, the renal and intestinal circulations changed most dramatically. In addition, we found that EDNO synthesis inhibition caused no changes in either blood pH or p^CO2and only a slight increase in p^Osub 2. This indicates that, in conscious rats, EDNO is an important regulator of blood flow, and the hemodynamic changes are not related to changes in blood gases.

It has been reported that, in barbiturate-anesthetized animals, as in conscious rats, EDNO synthesis inhibition increases BP and TPR while decreasing CO and HR. [10,12,24] Loeb and Longnecker [10] demonstrated that L-NMMA increased vascular resistance from 45% in the cerebral circulation to more than 200% in brown fat. Again, the current study confirms the heterogeneous response of different vascular beds to EDNO synthesis inhibition in barbiturate-anesthetized rats. Although L-NAME tended to increase resistance in all vascular beds, blood flow to the brain and heart were not changed by L-NAME, consistent with the study by Wang et al. [12] Thus, blood flow to these beds may be under more profound central control and, therefore, sensitive to the general influence of anesthesia. Under such conditions, suppression of the central nervous system (by anesthesia) will reduce the effects of other regulatory systems, such as EDNO.

Halothane exerts a variety of influences on systemic and regional hemodynamics. It has been shown to decrease BP by decreasing cardiac output without altering systemic vascular resistance. [17] Halothane has also been shown to alter basal blood flow to a variety of vascular beds; it increased cerebral blood flow while decreasing blood flow to the heart, kidney, liver, and muscle. [13,16,17] Similar to earlier reports, we found that halothane decreased basal BP, CO, and HR while having no significant effect on TPR compared with inactin-anesthetized rats. Moreover, halothane tended to decrease basal blood flow and increase vascular resistance in the heart, kidney, and intestine, but it increased blood flow to the brain. Thus, halothane tends to increase basal resistance, particularly in the visceral beds. This is similar to the effects seen with EDNO synthesis inhibition, except that there is no increase in systemic pressure, presumably because of the bradycardia and decreased cardiac output after halothane, Wang et al. [14] reported that, during halothane anesthesia, EDNO synthesis inhibition did not increase BP, indicating that the influence of EDNO on systemic pressure and TPR is largely diminished by halothane. However, Greenblatt et al. [15] observed that, in rats anesthetized with either halothane or isoflurane, L-NMMA induced a pressor response, indicating that EDNO was still a major determinant of regional blood flow and TPR. Our findings agree with Wang [14] in that the systemic hemodynamic response to L-NAME was largely attenuated in halothane-anesthetized rats compared with either conscious or barbiturate-anesthetized rats. In addition, the hemodynamic response of peripheral vascular beds to L-NAME was markedly attenuated. Our findings indicate that in vivo halothane greatly diminishes or eliminates the influence of EDNO on systemic and regional hemodynamics compared with either conscious or inactin-anesthetized rats. More recently, Wang et al. [25] reported that halothane anesthesia eliminates the systemic pressor response to two different EDNO synthesis inhibitors, but not to exogenous vasoconstrictors, such as angiotensin II or norepinephrine. As suggested previously, [17] these authors also determined that the lack of a pressor response was caused by a marked attenuation of the increase in TPR. As in our study, they found that halothane diminished the local increase in vascular resistance in a variety of organ beds, except for the liver and spleen. They concluded that halothane suppresses the constrictor effect of EDNO synthesis inhibitors by releasing NO or potentiating its vasodilator action. Although our results are similar, we interpret them quite differently. If the hemodynamic responses to L-NAME reflect the contribution of EDNO to systemic or local vascular tone, dissipating these responses indicates that halothane directly or indirectly blocks either the synthesis of EDNO or its targets, rather than potentiating its effect. Clearly, if halothane potentiated EDNO, blocking its synthesis and removing its influence would amplify, rather than eliminate, systemic or local hemodynamic responses.

Both barbiturate and halothane anesthesia (without ventilation) resulted in similar decreases in the p^O2, and these were not changed further by EDNO synthesis inhibition. Because the response to L-NAME was so different with halothane compared with inactin, despite having similar p^O2, these findings indicate that anesthesia-induced changes in arterial blood gases cannot account for the different responses to EDNO synthesis reported, nor the apparent lack of influence of L-NAME on systemic and regional tone in halothane-anesthetized rats. It should be noted that, in anesthetized rats breathing room air, a p^O2of 60–70 represents a significant desaturation of arterial hemoglobin and, therefore, may not represent what would be observed in a controlled clinical setting.

The mechanism by which halothane interferes with the influence of EDNO on vascular tone is not yet defined. Several possibilities have emerged in the recent literature. Both in vivo and in vitro studies have demonstrated that volatile anesthetics, including halothane, may cause a marked decrease in the activity of nitric oxide synthase [26] or of soluble guanylate cyclase, [27,28] the second messenger target of EDNO in vascular smooth muscle. Interruption of either of these steps in the cascade of endothelium-derived relaxation could account for the halothane-mediated decreases in basal cGMP levels. [29] Some volatile anesthetics may inhibit EDNO by generating superoxide anions, closely related oxygen-free radicals, [19] which are known inhibitors of EDNO, or both. [30,31] Finally, it has been demonstrated that halothane increases calcium efflux and decreases calcium influx through receptor-operated channels in vascular smooth muscle. [32] Therefore, halothane could interfere with the action of EDNO by altering calcium fluxes, because calcium is a cofactor for EDNO synthesis.

Our studies demonstrate that, in both conscious and barbiturate-anesthetized rats, EDNO is an important regulator of systemic and regional hemodynamics. Also, although the response to EDNO synthesis inhibition was qualitatively similar in both conscious and barbiturate-anesthetized rats, the changes were generally greater in conscious animals, particularly in the cardiac and cerebral circulations. By comparison, halothane decreased BP and CO and virtually eliminated the systemic and regional hemodynamic responses to inhibition of EDNO synthesis. These observations indicate that halothane has some negative influence over the phenomenon of endothelium-derived vasodilation, either directly on synthesis of EDNO or indirectly on some step in its physiologic pathway. The implications for an anesthetic agent suppressing or eliminating an important regulator of vascular tone in certain patients, especially those undergoing vascular surgery, may not yet be appreciated. However, it is clear that the efficacy or appreciation of certain anesthetics in different types of procedures may be influenced by how these agents interact with the production or the effect of nitric oxide from the endothelium.