Lidocaine Induces a Reversible Decrease in Alveolar Epithelial Fluid Clearance in Rats

Male Sprague-Dawley rats weighing 300–350 g were studied. The rats were housed in air-filtered, temperature-controlled rooms. The University of California–San Francisco Animal Research Committee approved the protocol for these studies.

The rats were anesthetized with pentobarbital (60 mg/kg administered intraperitoneally), and anesthesia was maintained with 30 mg/kg body weight of pentobarbital sodium every 1 h. An endotracheal tube (PE-220; Clay Adams, A Becton-Dickinson Company, Parsippany, NJ) was inserted through a tracheotomy. Pancuronium bromide (0.3 mg · kg body weight⁻¹· h⁻¹; Pavulon^®; Organon Inc., West Orange, NJ) was given for neuromuscular blockade. A catheter (PE-50; Clay Adams) was inserted into the carotid artery to monitor systemic arterial pressure and to obtain blood samples. A second catheter was inserted into the jugular vein for intravenous drug administration. The rats were maintained in the right lateral decubitus position during the experiments to facilitate fluid deposition into the right lung. They were ventilated with a constant-volume pump (Harvard Apparatus, Dover, MA) with an inspired oxygen fraction of 1.0 and peak airway pressures of 8–12 cm H²O, supplemented with positive end-expiratory pressure of 3 cm H²O. The respiratory rate was adjusted to maintain the arterial carbon dioxide tension between 35 and 40 mmHg. Body temperature was kept constant at 38°C with a thermostatically controlled pad.

A 5% albumin (bovine serum albumin; Sigma Biochemical Co., St. Louis, MO) solution was prepared using Ringer lactate and adjusted with NaCl to be isosmolar with the rat's circulating plasma, as in our previous studies. 2,8Anhydrous Evan blue dye (0.5 mg; Aldrich Chemical Company Inc., Milwaukee, WI) was added to the albumin solution to confirm the location of the instillate at the end of the study, and 1 μCi of ¹²⁵I-labeled human serum albumin (Frosst Laboratories, Montreal, Quebec, Canada) was added to the albumin solution to measure the protein permeability of the alveolar epithelium (see below). A sample of the instilled solution was saved for total protein measurement to be used to calculate alveolar fluid clearance (see Measurements), radioactivity counts, and the water-to-dry weight ratio measurements so that the dry weight of the protein solution could be subtracted from the final lung water calculation.

In all experiments, after the surgery, heart rate and systemic blood pressure were allowed to stabilize for 1 h. Then a vascular tracer, 1 μCi of ¹³¹I-labeled human albumin (Frosst Laboratories), was injected intravenously to calculate the flux of the plasma protein into the lung interstitium. Thirty minutes later, the 5% albumin instillate solution with 1 μCi of ¹²⁵I-labeled albumin (alveolar tracer) was instilled into the right lower lobe to calculate the flux of protein from the air spaces into the circulating plasma. The instillate solution (3 ml/kg) was instilled over 20 min using a 1-ml syringe and a polypropylene tube (0.5 mm ID) advanced into a wedged position in the right lung. Mechanical ventilation was uninterrupted during instillation, and 1 h after instillation, the abdomen was opened and the rats were exsanguinated by transecting the abdominal aorta. Urine was obtained for radioactivity counts by puncturing the bladder. The lungs were removed through a median sternotomy. An alveolar fluid sample from the distal airspaces (0.1– 0.2 ml) was obtained by gently passing the sampling catheter (PE-50 catheter, 0.5 mm ID) into a wedged position in the instilled area of the right lower lobe and aspirating. After centrifugation of the samples, total protein concentration and radioactivity of the sampled fluid were measured. The right and left lungs were homogenized separately for the water-to-dry weight ratio measurements and radioactivity counts.

All rats were studied for 1 h and then killed and processed as described in the general experimental protocol.

In group 1 (n = 6), after the baseline period, 3 ml/kg body weight of the 5% albumin solution with ¹²⁵I-labeled human serum albumin were instilled into one lung.

To study the effects on intraalveolar lidocaine, in group 2 (n = 11) we added 10⁻⁵m lidocaine (Sigma Biochemicals) to the instilled 5% albumin solution (n = 5). In the studies of the effects of intravenous lidocaine (n = 6), the rats received a bolus of 2 mg/kg lidocaine intravenously followed by a continuous intravenous infusion at a rate of 2 mg · kg⁻¹· h⁻¹lidocaine until the end of the experiment. The intravenous lidocaine infusion started 30 min before the instillation of 5% albumin solution. The intravenous infusion at this dose results in a lidocaine concentration of 10⁻⁵m in the plasma, which is a clinically relevant concentration when lidocaine is used intravenously. 19After the baseline period, 3 ml/kg body weight of the 5% albumin solution was instilled into one lung.

To study the effects of β-adrenergic stimulation in the presence of lidocaine, in group 3 (n = 8) we instilled the rats with terbutaline as follows. Terbutaline (10⁻⁴m; Sigma) and lidocaine (10⁻⁵m) were added to the instilled 5% albumin instillate (n = 4). Terbutaline (10⁻⁴m; n = 4) without lidocaine was added to the 5% albumin instillate and instilled in a separate control group.

Normally, amiloride reduces alveolar fluid clearance in rats by 40–60%. 6To study the effect of lidocaine on amiloride inhibition, in group 4 (n = 8) we instilled the rats with amiloride as follows. Amiloride (10⁻³m; ICN Biochemicals, Inc., Costa Mesa, CA) and lidocaine (10⁻⁵m) were added to the instilled 5% albumin solution (n = 4). Amiloride (10⁻³m; n = 4) without lidocaine was added to the 5% albumin instillate and instilled in a separate control group.

We hypothesized that lidocaine may decrease sodium transport by inhibition of basolateral potassium channels. To test this hypothesis, in group 5 (n = 6) we used QX-314 (Sigma), a quaternary analog of lidocaine that blocks voltage-gated sodium channels and potassium channels, similar to the effects described for lidocaine. 13,14Since QX-314 is permanently positively charged, it is not able to cross the cell membranes. If QX-314 is administered intravenously, it can accumulate in the lung interstitium on the basal side of the alveolar epithelial cells; if QX-314 is administered into the alveolar space, it should remain on the apical side of the alveolar epithelium because the alveolar epithelium is very tight, 20thus making it difficult for QX-314 to cross the tight epithelial junctions. Therefore, theoretically QX-314 could inhibit basolateral potassium channels if given intravenously, but it does not have this effect when administered in the alveolar spaces.

To study the intraalveolar effects of QX-314, we added it (10⁻⁵m) to the instilled 5% albumin solution (n = 3). In the studies of the effect of intravenous QX-314 (n = 3), the rats received the same dose previously used with lidocaine, i.e. , a bolus 2 mg/kg QX-314 intravenous dose followed by a continuous intravenous infusion at a rate of 2 mg · kg⁻¹· h⁻¹, until the end of the experiment. The intravenous QX-314 infusion started 30 min before the instillation of 5% albumin solution.

Systemic arterial, central venous, and airway pressures were continuously measured. Arterial blood gases were measured at the beginning and the end of the studies. Samples from the instilled protein solution, from the final distal air space fluid, and from the initial and final blood were collected to measure total protein concentration with an automated analyzer (AA2 Technicon, Tarrytown, NY).

Two different methods were used to measure the bidirectional protein permeability across the lung endothelial and epithelial barriers. 3,8The first method measures residual ¹²⁵I-albumin (the air space protein tracer) in the lungs and the accumulation of ¹²⁵I-albumin in the plasma. The second method measures accumulation of ¹³¹I-albumin (the vascular protein tracer) in the air spaces of the lungs.

For measuring the flux of the alveolar tracer protein, ¹²⁵I-albumin, from the lung, several measurements were necessary. The total radioactivity instilled into the lungs (¹²⁵I-albumin, counts per minute per gram) was calculated by multiplying the radioactivity in aliquots of the instillate by the volume instilled. The residual quantity of ¹²⁵I-albumin in the lungs at the end of the experiment was calculated by the average of the duplicate radioactivity counts from the lung homogenates from the instilled lung multiplied by the volume of the lung homogenate. To measure the passage of ¹²⁵I-albumin from the alveolar spaces into the blood, the radioactivity counts in the plasma samples were multiplied by the plasma volume (milliliter). The plasma volume was calculated by the following relation:

To measure the accumulation of the vascular tracer protein, ¹³¹I-albumin, into the air spaces of the lungs, the concentration of ¹³¹I-albumin in the alveolar sample was measured at the end of the experiment and compared with the concentration in the plasma.

Changes in the concentration of the instilled nonlabeled bovine albumin over the study period (1 h) were used to measure fluid clearance from the distal air spaces, as we have in several previous experimental studies. 2,8An increase in alveolar protein concentration is a direct measurement of absorption of water from the instilled solution since very little protein entered or escaped from the alveolar space during the studies (see Results). Because some reabsorption may have occurred across distal bronchial epithelium, the term alveolar  does not imply that all fluid reabsorption occurred at the alveolar level.

To determine ¹²⁵I binding to albumin, precipitation of all protein in the samples was performed by adding 20% trichloroacetic acid to all tubes. The tubes were then centrifuged to obtain the supernatant for measurement of free ¹²⁵I radioactivity. The results are expressed as a percentage of the unbound ¹²⁵I radioactivity to the total amount of ¹²⁵I-albumin radioactivity instilled. All fluid samples always had less than 1% of unbound activity.

To further test the hypothesis that lidocaine acts on the basolateral surface of the alveolar epithelium, we evaluated the effect of lidocaine on the ENaC, a channel that is expressed only on the apical surface of alveolar epithelial cells. 21To test the hypothesis that lidocaine did not act on ENaC, the following studies were conducted. Because oocyte expression studies suggest that all three subunits (α, β, γ) are required for optimal function of the channel, 22 Xenopus  oocytes were transfected with all three human subunits of ENaC.

Mature female Xenopus laevis  were maintained at 20°C with a 12-h light–dark cycle. Individual females were anesthetized in ice, and oocyte clusters were surgically removed from the ovary. Oocyte clusters were torn apart with forceps in ND96 medium containing 96 mm NaCl, 2 mm KCl, 10 mm HEPES, and 1.8 mm CaCl^2,at pH 7.4. Denuded oocytes were obtained by collagenase digestion (type IA, Sigma, 370 U/ml) during 2 h at room temperature and rinsed several times in ND96. Stage 5–6 oocytes were selected and incubated overnight at 18°C in ND96 medium with gentamycin (50 mg/ml). Healthy oocytes were then selected and injected with 50 nl cRNA (50 ng/ml). The oocytes were incubated for 1 day after the injection in ND96 medium supplemented with 50 mg/ml gentamycin, 10 μm amiloride, and 5% inactivated horse serum. For the measurements in this study, the oocytes were punctured with two conventional microelectrodes (3 mol/1 KCl) for voltage clamp experiments (TEV200 voltage clamp; Dagan Corporation, Minneapolis, MN). The specific ENaC current was defined as the total current recorded at −70 mV holding potential minus the current recorded at the same holding potential in the presence of 20 μm amiloride. Recordings were performed in ND96, where CaCl²was replaced by 1.8 mm BaCl²to block endogenous oocyte K⁺conductance.

All data are summarized as mean ± SE. Experimental and control groups were compared using one-way analysis of variance if significant by post hoc  Sheffé test. P < 0.05 was considered statistically significant.

The mean systemic arterial pressure, peak airway pressure, and arterial blood gases were similar at baseline in all groups. There was no effect of intraalveolar or intravenous lidocaine on mean systemic arterial pressure or oxygenation in rats (table 1).

Alveolar lidocaine (10⁻⁵m) decreased alveolar fluid clearance over 1 h by 50% compared with control rats (fig. 1). There were similar effects with intravenous lidocaine (plasma concentration, 10⁻⁵m) on alveolar fluid clearance over 1 h (fig. 1).

The β-adrenergic agonist, terbutaline, increased alveolar fluid clearance in control rats by 30% (fig. 2). Interestingly, terbutaline overcame the effect of lidocaine, resulting in the normal up-regulation of alveolar fluid clearance with β²-agonist stimulation (fig. 2).

Amiloride (10⁻⁴m) inhibited alveolar fluid clearance in control rats by approximately 60%. When amiloride was added to the instilled albumin solution with lidocaine, there was no additional inhibition of alveolar fluid clearance (fig. 3).

When QX-314 (10⁻⁵m) was added to the instilled albumin solution, there was no effect, similar to the control studies. By contrast, intravenous QX-314 decreased alveolar fluid clearance over 1 h by 50%, similar to the effect of intravenous lidocaine (fig. 4).

There was no effect of intravenous or intraalveolar lidocaine on either the movement of the vascular tracer (¹³¹I-albumin) into the air spaces of the lung or the alveolar tracer (¹²⁵I-albumin) from the lung to the plasma compared with the controls. Thus, there was no change in epithelial or endothelial permeability to protein in the lung (table 2).

Lidocaine had no direct effects on the ENaC. This result was demonstrated after heterologous expression of the three human ENaC subunits in Xenopus  oocytes. A large amiloride-sensitive current developed in the oocytes 1 day after injection, and this current was not altered by 10 μm lidocaine (fig. 5).

There were several clinical reasons for studying the effect of lidocaine on alveolar epithelial fluid clearance in the lung. First, lidocaine has been reported to impair ion transport in nasal and tracheal epithelium. 10,11If this effect occurred at the level of the alveolar or distal airway epithelium, then lidocaine could either lower the threshold for developing alveolar edema or slow the resolution of edema because active ion transport is the primary mechanism responsible for the resolution of alveolar edema. 2,3,6,8Interestingly, recent clinical studies have implicated lidocaine as a possible cause of pulmonary edema in patients undergoing liposuction surgery. 17,18,23In addition, because lidocaine is widely used as an antiarrhythmic agent in patients with cardiac disease who are susceptible to the development of pulmonary edema, it is important to know if lidocaine can impair alveolar epithelial fluid transport. Also, the postoperative use of lidocaine in patients with acute cardiac disease could lower the threshold for developing pulmonary edema.

The principal results of this study can be summarized as follows. Lidocaine decreased baseline alveolar epithelial fluid clearance by 50%. This effect occurred whether lidocaine was administered intravenously in a clinically relevant dose or directly in the distal airspaces of the lung. β-adrenergic stimulation with terbutaline was successful in overcoming the lidocaine effect and restoring alveolar fluid clearance to the normal up-regulated level. 2,6There may be several mechanisms by which lidocaine could inhibit alveolar epithelial fluid clearance.

First, does lidocaine act directly on transepithelial sodium transport? It is well accepted that alveolar fluid clearance is driven, in part, by amiloride-sensitive and -insensitive sodium transport. 6,21Because sodium transport is an important driving force, and lidocaine can induce a blockade of voltage-gated channels through interference with the conformational changes that underlie the activation process, 12it is possible that inhibition of sodium channels would be one mechanism. There are sodium channels present in the alveolar epithelium that may be candidates for inhibition by lidocaine. The most likely candidate on the apical surface would be the ENaC. However, when lidocaine was instilled into the lungs together with amiloride, no additional inhibition occurred. This result suggested that lidocaine and amiloride might compete for the same inhibitory site on ENaC. A second possibility would be that lidocaine inhibited an accessory sodium transport pathway that required amiloride inhibition, a theory that could explain why amiloride did not induce a further decrease of alveolar liquid clearance in the presence of lidocaine.

Therefore, experiments to investigate the first hypothesis were conducted. First, we evaluated the effect of lidocaine on the function of ENaC expressed in Xenopus  oocytes. In these studies, lidocaine had no direct effects on the function of ENaC. To investigate the second hypothesis, we used a permanent polarized analog of lidocaine, QX-314, and administered this compound either intravenously or into the distal air spaces of the lung. The decrease in alveolar fluid clearance occurred only with intravenous administration of QX-314 and not when QX-314 was administered into the distal airspaces of the lung. This result supports the hypothesis that the primary effect of lidocaine occurs from its action on the basal side of the alveolar epithelium. Thus, the lidocaine-induced decrease in alveolar fluid clearance apparently was not the result of a decrease in sodium channel apical uptake since intraalveolar QX-314 did not decrease alveolar fluid clearance and because there was no direct effect of lidocaine on ENaC activity in the oocytes. Thus, the effect of lidocaine must be explained by another mechanism.

Could lidocaine exert its effect by inhibiting the basolateral NaKATPase or potassium channels, or by both mechanisms? Because QX-314 was able to reduce the alveolar fluid clearance intravenously but not in alveolar spaces, these data suggested that lidocaine acts primarily on the basal side of the alveolar epithelium. This inhibition could have been the result of either a direct effect on the NaKATPase activity or a secondary inhibition of potassium channels. Theoretically, lidocaine may have altered the activity of NaKATPase, although there is no previous report that lidocaine has this effect. On the other hand, it is well known that lidocaine can inhibit basolateral K⁺channels in renal brush-border membranes, 24urinary bladder epithelia, 25,26turtle colon epithelial cells, 27and frog alveolar epithelium. 14Our studies do not resolve whether the effect of lidocaine in these in vivo  studies is mediated by a reduction in the activity of NaKATPase or inhibition of K⁺channels, or both. However, we favor the latter explanation, that lidocaine inhibits basal potassium channels, because of previous in vitro  work demonstrating this effect. 14,28In addition, a previously published study from our research group demonstrated that a potassium channel opener (YM934) increased the rate of alveolar fluid clearance in an ex vivo  human lung preparation. 29Thus, in the current study, lidocaine may well have had the opposite effect on potassium channels.

How would these mechanisms explain the lidocaine-induced decrease in vectorial salt and water transport across the alveolar epithelium? In tight epithelia there is a relation between ion transport of the apical membrane sodium-selective uptake and the basolateral membrane potassium-selective extrusion. Normally, an increase of sodium uptake across the apical membrane leads to an increased extrusion of sodium at the basolateral membrane by the NaKATPase to maintain a constant low intracellular sodium concentration. This effect is associated with an increase in basal membrane conductance of potassium channels. In urinary bladder, inhibition of apical sodium channels by amiloride results in a decrease in basolateral membrane potassium permeability. Conversely, the blockade of basolateral potassium channels leads to a decrease in apical sodium conductance. 30 

Why was terbutaline able to reverse most of the inhibitory effect of lidocaine? It is well known that β-adrenergic agonists, such as terbutaline or isoproterenol, can increase the activity of NaKATPase in alveolar type II cells. 31,32NaKATPase activity enhances extrusion of sodium, resulting in enhanced apical sodium uptake. To maintain the cell potential, potassium needs to be extruded by basolateral potassium channels. In the presence of lidocaine, this extrusion is still possible because different subtypes of basolateral potassium channels exist and only some of them are inhibited by lidocaine. 33The blockade of all basolateral potassium channels would antagonize the stimulatory effect of terbutaline on alveolar fluid transport similar to the effect reported in fetal distal lung epithelium in vitro  with barium, a nonspecific potassium channel blocker. 34 

What is the clinical significance of these results? Whatever the mechanisms responsible for the formation of pulmonary edema, a decrease in the alveolar epithelial fluid transport capacity can delay or impair the resolution of alveolar edema. We found that in critically ill patients ventilated for acute respiratory distress syndrome, the patients who were able to maximize alveolar fluid clearance had a better outcome. 35,36Critically ill patients with cardiac arrhythmias receive the same dose of lidocaine intravenously (2 mg · kg⁻¹· h⁻¹) that was administered in this experimental study. Many of these patients are predisposed to develop hydrostatic pulmonary edema because of acute and chronic cardiac disease. Conceivably, lidocaine therapy may lower the threshold for the development of alveolar edema, especially in the presence of elevated left atrial pressure. In addition, there have been several reports of pulmonary edema that occurred after plastic surgery, especially liposuction, concomitant with the use of high doses of lidocaine along with fluid loading. 16,18In the presence of volume loading and a lidocaine-induced reduction in cardiac contractility, inhibition of alveolar epithelial fluid transport could contribute to or worsen the development of pulmonary edema.

In summary, clinically relevant concentrations of lidocaine reduced alveolar epithelial fluid transport in rats. The inhibitory effect of lidocaine was probably mediated by inhibition of potassium channels or NaKATPase activity on alveolar epithelium. The inhibitory effect of lidocaine was overcome with β-adrenergic receptor stimulation. This reversibility may be particularly relevant to patients with cardiac failure who are often treated with dobutamine, an agent that can augment alveolar fluid clearance in rats by stimulation of β²-adrenergic receptors. 37