Volatile anesthetics protect against cardiac ischemia-reperfusion injury via adenosine triphosphate-dependent potassium channel activation. The authors questioned whether volatile anesthetics can also protect against renal ischemia-reperfusion injury and, if so, whether cellular adenosine triphosphate-dependent potassium channels, antiinflammatory effects of volatile anesthetics, or both are involved.
Rats were anesthetized with equipotent doses of volatile anesthetics (desflurane, halothane, isoflurane, or sevoflurane) or injectable anesthetics (pentobarbital or ketamine) and subjected to 45 min of renal ischemia and 3 h of reperfusion during anesthesia.
Rats treated with volatile anesthetics had lower plasma creatinine and reduced renal necrosis 24-72 h after injury compared with rats anesthetized with pentobarbital or ketamine. Twenty-four hours after injury, sevoflurane-, isoflurane-, or halothane-treated rats had creatinine (+/- SD) of 2.3 +/- 0.7 mg/dl (n = 12), 1.8 +/- 0.5 mg/dl (n = 6), and 2.4 +/- 1.2 mg/dl (n = 6), respectively, compared with rats treated with pentobarbital (5.8 +/- 1.2 mg/dl, n = 9) or ketamine (4.6 +/- 1.2 mg/dl, n = 8). Among the volatile anesthetics, desflurane demonstrated the least reduction in plasma creatinine after 24 h (4.1 +/- 0.8 mg/dl, n = 12). Renal cortices from volatile anesthetic-treated rats demonstrated reduced expression of intercellular adhesion molecule 1 protein and messenger RNA as well as messenger RNAs encoding proinflammatory cytokines and chemokines. Volatile anesthetic treatment reduced renal cortex myeloperoxidase activity and reduced nuclear translocation of proinflammatory nuclear factor kappaB. Adenosine triphosphate-dependent potassium channels are not involved in sevoflurane-mediated renal protection because glibenclamide did not block renal protection (creatinine: 2.4 +/- 0.4 mg/dl, n = 3).
Some volatile anesthetics confer profound protection against renal ischemia-reperfusion injury compared with pentobarbital or ketamine anesthesia by attenuating inflammation. These findings may have significant clinical implications for anesthesiologists regarding the choice of volatile anesthetic agents in patients subjected to perioperative renal ischemia.
RENAL dysfunction secondary to ischemia–reperfusion injury is a major clinical concern in anesthetized patients undergoing surgery involving the kidney or aorta.1,2The risk of acute renal failure is significantly increased in patients with impaired preoperative renal function.3Millions of patients are exposed to inhalational anesthetics annually. Besides their anesthetic effects, volatile anesthetics have significant nonanesthetic physiologic effects. Recent evidence exists indicating that volatile anesthetics protect against ischemia–reperfusion injury in the heart4–6and lung.7–9Specifically, pretreatment with volatile anesthetics before cardiac ischemia protects against ischemia–reperfusion injury.4–6The mechanisms of organ protection by volatile anesthetics are unclear; however, several studies have suggested that volatile anesthetics protect the heart via activation of adenosine triphosphate–dependent potassium (K+ATP) channels. Other studies suggest that volatile anesthetics protect against ischemia–reperfusion injury in the heart10,11and lung12,13 via antiinflammatory effects. However, renal protective effects of volatile anesthetics have not been reported.
A significant component of renal tubular injury occurs as a result of the inflammatory response during the reperfusion period. Reperfusion after ischemic injury triggers activation of several transcription factors, including nuclear factor (NF)-κB, which in turn alter the transcription of multiple genes associated with the inflammatory response, including intercellular adhesion molecule (ICAM)-1, interleukin (IL)-8, and tumor necrosis factor (TNF)-α.14–16Induction of proinflammatory chemokines and cytokines after ischemia and reperfusion has been increasingly implicated in renal injury.14,17Several key cytokines (e.g ., TNF-α) as well as chemokines (monocyte chemotactic protein [MCP]-1, macrophage inflammatory protein [MIP]-2, IL-8, and interferon γ–inducible protein [IP]-10) are toxic to renal tubules and promote attraction of leukocytes to the injured site.
We aimed to determine whether several commonly used volatile anesthetics protect against renal ischemia–reperfusion injury to a similar extent using an in vivo rat model. Second, we questioned whether volatile anesthetics decreased renal dysfunction induced by ischemia–reperfusion injury by mechanisms involving the activation of K+ATPchannels, reducing inflammation, or both.
Materials and Methods
Pentobarbital and ketamine were purchased from Henry Schein Veterinary Co. (Indianapolis, IN). Glibenclamide was purchased from Sigma (St. Louis, MO). Isoflurane and sevoflurane were obtained from Abbott Laboratories (North Chicago, IL). Desflurane was obtained from Baxter Healthcare Corp. (Deerfield, IL). Halothane was obtained from Halocarbon Laboratories (River Edge, NJ).
All animal protocols were approved by the Institutional Animal Care and Use Committee of Columbia University (New York, New York). Adult male Sprague-Dawley rats (weight, 250–300 g) were anesthetized with intraperitoneal pentobarbital (50 mg/kg body weight, or to effect), ketamine (50 mg/kg body weight, or to effect), or volatile anesthetics (approximately 1 minimum alveolar concentration [MAC]; 1.2% isoflurane, 2.2% sevoflurane, 0.8% halothane, or 6.7% desflurane). Pentobarbital- and ketamine-anesthetized rats were allowed to breathe room air spontaneously on an electric heating pad under a warming light, whereas volatile anesthetic–treated rats breathed spontaneously while receiving approximately 1 MAC volatile agent in room air as described below. Body temperature was monitored with a rectal probe and maintained at 37°C. In some animals, the bladder and right femoral vein and artery were cannulated with heparinized (10 U/ml) polyethylene tubing for the measurement of blood pressure and renal blood flow.
To initially anesthetize rats with volatile anesthetics (isoflurane, sevoflurane, halothane, or desflurane), rats were placed in an airtight 10-l chamber on a warming blanket with inflow and outflow hoses located at the top and bottom of the chamber, respectively. The volatile anesthetics were delivered in room air at 5 l/min using agent specific vaporizers. The vaporizers were set to maintain chamber volatile anesthetic concentrations at 1 MAC monitored by infrared analyzer sampling gas at the outflow hose. In studies where volatile anesthetic administration was maintained through the ischemia and reperfusion period, the animals were removed from the chamber during anesthesia and allowed to breathe identical anesthetic concentrations through a nose cone connected in parallel to the gas chamber. Each animal was subjected to midline laparotomy, right nephrectomy and sham operation, or 45 min left renal ischemia during anesthesia. After 45 min of left renal ischemia, occlusion clips were removed, and the abdomen was closed in two layers. Each animal was returned to the chamber and allowed to breathe identical anesthetic concentrations spontaneously for an additional 3 h. Pentobarbital- and ketamine-treated animals were returned to their cages to recover from anesthesia.
In separate experiments, we tested whether a 1-h pretreatment with a volatile anesthetic (sevoflurane, desflurane, isoflurane, or halothane) provided protection against renal ischemia–reperfusion injury during intraperitoneal pentobarbital anesthesia. Rats were anesthetized with volatile anesthetics for 1 h. After 1 h of volatile anesthetic treatment, they were allowed to awaken and were immediately anesthetized with intraperitoneal pentobarbital for a midline laparotomy. In preliminary experiments, recovery from volatile anesthetic was rapid, and the animals tolerated additional pentobarbital anesthesia well.
To determine whether volatile anesthetic–mediated renal protection involves activation of K+ATPchannels, rats were pretreated with 6 mg/kg intravenous glibenclamide, a K+ATPchannel antagonist, 30 min before sevoflurane anesthesia. We have demonstrated previously that this dose of glibenclamide effectively blocks K+ATPchannels in rats.18
Estimation of Renal Blood Flow
We determined the effects of anesthetics on renal blood flow. Sevoflurane was chosen as a representative volatile anesthetic. Rats were anesthetized with pentobarbital, ketamine, or sevoflurane (2.2% or approximately 1 MAC) and preischemic and postischemic renal blood flows were determined using para-aminohippuric acid clearance measurements. In brief, after right nephrectomy, the femoral vein and bladder were cannulated for periodic sampling of blood and urine for para-aminohippuric acid. After a bolus of para-aminohippuric acid (8 mg/kg) and 1 h of continuous infusion of para-aminohippuric acid (5.8 mg/ml at 0.055 ml/min), plasma and urine samples were collected immediately before renal ischemia and 60 min after renal ischemia for spectrophotometric para-aminohippuric acid detection.19The clearance of para-aminohippuric acid and renal blood flow were calculated from standard equations.
Measurement of Plasma Creatinine
Plasma creatinine was measured by using a commercially available colorimetric method (Sigma). Plasma samples were obtained from the tail vein 24, 48, and 72 h after ischemic injury.
Histologic Examinations and Detection of Necrosis
Some animals were killed 24 h after injury for histologic examination. For histologic preparation, explanted kidneys were bisected along the long axis and cut into three equal-sized slices. Kidney slices were obtained from rats undergoing sham operation during pentobarbital anesthesia or renal ischemia–reperfusion injury during pentobarbital, ketamine, or volatile anesthesia and were fixed in 10% formalin overnight. After automated dehydration through a graded alcohol series, transverse kidney slices were embedded in paraffin, sectioned at 5 μm, and stained with hematoxylin–eosin. Morphologic assessment was performed by a renal pathologist (S. H. N.) who was unaware of the treatment that the animal had received. A grading scale of 0–4, as outlined by Jablonski et al .,20was used for the histopathologic assessment of ischemia-reperfusion–induced damage of the proximal tubules.
Semiquantitative Reverse Transcriptase Polymerase Chain Reaction for Proinflammatory Cytokines
Twenty-four hours after renal ischemic injury, renal cortices were dissected, and total RNA was extracted using Trizol (Invitrogen, Carlsbad, CA) reagent. RNA concentrations were determined with spectrophotometric readings at 260 nm and run on agarose gels to verify equal RNA input and RNA quality. Reverse-transcriptase polymerase chain reaction was performed to analyze the expression of proinflammatory (TNF-α, ICAM-1, MCP-1, MIP-2, IL-8, and IP-10) genes. Primers were designed based on published GenBank sequences (Bethesda, MD) for the rat (table 1). Primer pairs were chosen to yield expected polymerase chain reaction products of 200–600 base pairs and to amplify a genomic region that spans one or two introns to eliminate the confounding effect of amplifying contaminating genomic DNA. Primers were purchased from Sigma Genosys (The Woodlands, TX). Reverse-transcriptase polymerase chain reaction was performed using the Access Reverse Transcriptase Polymerase Chain Reaction System (Promega, Madison, WI), which is designed for a single tube reaction for first-strand complementary DNA synthesis (48°C for 45 min) using avian myeloblastosis virus reverse transcriptase and subsequent polymerase chain reaction using Tfl DNA polymerase. Polymerase chain reaction cycles included a denaturation step of 94°C for 30 s followed by an optimized annealing temperature (table 1) for 1 min followed by a 1-min extension period at 68°C. All polymerase chain reaction reactions were completed with a 7-min incubation at 68°C to allow enzymatic completion of incomplete complementary DNAs. The polymerase chain reaction cycle number for each primer pair was optimized to yield linear increases in the densitometric measurements of resulting bands with increasing cycles of polymerase chain reaction (15–30 cycles). The starting amount of RNA was also optimized to yield linear increases in the densitometric measurements of resulting bands at an established number of polymerase chain reaction cycles. Based on these preliminary experiments, 0.5–1.0 μg total RNA was used as template for all reverse-transcriptase polymerase chain reactions. The number of polymerase chain reactions cycles yielding linear results was determined in preliminary studies. For each experiment, we also performed semiquantitative reverse-transcriptase polymerase chain reactions under conditions yielding linear results for glyceraldehyde-3-P dehydrogenase (15 cycles) to confirm equal RNA input. The products were resolved in 6% polyacrylamide gel and stained with Syber Green (Roche, Indianapolis, IN), and the band intensities were quantified using a Fluor-S Multi Imager (Biorad, Hercules, CA).
Renal Cortical Myeloperoxidase Assay
Myeloperoxidase is an enzyme present in leukocytes and is an index of tissue leukocyte infiltration.21Twenty-four hours after renal ischemic injury, renal cortex (approximately 200 mg) was dissected and homogenized for 30 s in 2 ml potassium phosphate buffer, 50 mm, pH 7.4, at 4°C. The samples were centrifuged for 15 min at 16,000g at 4°C, and the resultant pellet was resuspended in 2 ml potassium phosphate buffer, 50 mm, pH 7.4, with 0.5% hexadecyltrimethyl ammonium bromide at 4°C. The samples were sonicated for 30 s and centrifuged at 16,000g for 15 min at 4°C. Fifty microliters supernatant was mixed with 750 μl potassium phosphate buffer, 45 mm, pH 6.0, containing 0.167 mg/ml o-dianisidine and 0.3% hydrogen peroxide. Absorbance (460 nm) was measured over a period of 5 min (unit of enzyme activity = delta optical density/min/mg protein), and the relative myeloperoxidase activity was expressed as a percent of the sham-operated group. The remaining supernatant was used to determine protein concentrations.
Rat kidney cortical tissues were obtained at 24 h from rats subjected to sham operation during pentobarbital anesthesia or renal ischemia–reperfusion during pentobarbital or volatile anesthetic anesthesia and were dissected on ice, placed in ice-cold radioimmunoprecipitation buffer (150 mm NaCl, 50 mm Tris-HCl, 1 mm ethylenediamine tetra-acetic acid, 1% Triton-X, pH = 7.4), and homogenized for 10 s. The samples were then centrifuged for 10 min at 1,000g , and the resulting supernatant was collected, quantified (for protein concentration), and mixed to a 1× final concentration with Laemmli’s loading buffer (50 mm Tris-HCl, 1% 2-mercaptoethanol, 2% sodium dodecyl sulfate, 0.1% bromo-phenol blue, 10% glycerol). Equal amounts of protein (30 μg) were subjected to electrophoresis through a 7.5% polyacrylamide gel and transferred to polyvinylidene difluoride membranes. ICAM-1 expression was subsequently detected by immunoblotting using monoclonal antibody (sc-8439) (Santa Cruz Biotechnologies, Santa Cruz, CA) diluted 1:500 as described previously.22
Electrophoretic Mobility Shift Assay
Some animals were killed 6 h after sham operation or ischemia–reperfusion injury (during pentobarbital or sevoflurane anesthesia) for nuclear protein extraction. Renal cortices were dissected and immersed in 500 μl buffer A (10 mm 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid, pH 7.9, 10 mm KCl, 1.5 mm MgCl2, 20% glycerol, 0.2 mm phenylmethylsulfonyl fluoride, 0.5 mm dithiothreitol, protease inhibitor cocktail (Mini-complete–ethylenediamine tetra-acetic acid) (Roche)) for 10 min at 4°C. The cells were homogenized using a polytron homogenizer for 5 s to release the nuclei into solution and centrifuged at 18,000g for 5 min at 4°C. The supernatant was discarded, and the pellet was resuspended in 50 μl buffer B (20 mm 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid, pH 7.9, 1.5 mm MgCl2, 0.5 mm ethylenediamine tetra-acetic acid, 25% glycerol, 0.1% Triton X-100, 0.2 mm phenylmethylsulfonyl fluoride, 0.5 mm dithiothreitol, protease inhibitor cocktail) and incubated for 1 h at 4°C with occasional swirling to extract nuclear protein. The nuclei were centrifuged at 16,000g for 15 min, and the supernatant containing nuclear protein was used for electrophoretic mobility shift assay for NF-κB.
Electrophoretic mobility shift assay was performed using the Gel Shift Assay Systems (Promega, Winsconsin, WI). The oligonucleotides for NF-κB (Promega) consensus sequences were end-labeled with 10 μCi 32P-γ-ATP (Perkin Elmer Life Technology, Wellesley, MA) and purified using a G-25 spin column (Amersham Biosciences, Piscataway, NJ). Ten micrograms of the nuclear extract was incubated with 1 μl of the labeled probe for 20 min at room temperature and electrophoresed on a 4% polyacrylamide gel (200 V at 4°C). HeLa cell nuclear extract (Promega), 2 μg, was used for a positive control. One hundred–fold concentration of cold probe was coincubated as a competitor in a negative control reaction. In addition to the protein assay, we performed immunoblotting for histones to verify equal loading of nuclear fractions (data not shown). The gel was then transferred to blotting paper and exposed to film or scanned with a Phospho Imager (Molecular Dynamics, Piscataway, NJ).
Renal Cortical TNF-α Enzyme-linked Immunosorbent Assay
Tumor necrosis factor α concentrations were measured in homogenized whole rat kidney cortices harvested 24 h after ischemia–reperfusion injury from rats undergoing sham operation under pentobarbital anesthesia, renal ischemia–reperfusion under pentobarbital anesthesia, or renal ischemia–reperfusion under volatile anesthetics by enzyme-linked immunosorbent assay (R&D Systems, Minneapolis, MN).
Protein content was determined with the Pierce Chemical (Rockford, IL) bicinchoninic acid protein assay reagent with bovine serum albumin as a standard.
A one-way analysis of variance was used to compare mean values across multiple treatment groups with a Tukey post hoc multiple comparison test (e.g ., sham vs . ischemia–reperfusion, sevoflurane vs . desflurane). The ordinal values of the Jablonski scale were analyzed by the Kruskal-Wallis nonparametric test with Dunn posttest comparison between groups. In all cases, a probability statistic less than 0.05 was taken to indicate significance. All data are expressed throughout the text as mean ± 1 SD.
Twenty-four hours after renal injury, rats treated with 2.2% sevoflurane, 1.2% isoflurane, or 0.8% halothane during renal ischemia and 3 h of reperfusion had significantly lower plasma creatinine compared with rats anesthetized with pentobarbital or ketamine and subjected to renal ischemia–reperfusion injury (fig. 1A). Rats anesthetized with 1 MAC desflurane (6.7%) and subjected to 45 min of renal ischemic injury and 3 h of desflurane anesthesia also had lower plasma creatinine than rats anesthetized with pentobarbital, but remarkably, desflurane-treated rats had significantly higher plasma creatinine than rats anesthetized with other volatile anesthetics (fig. 1A). In contrast, 1 h of volatile anesthetic pretreatment did not protect renal function against ischemia–reperfusion injury (fig. 1B). Forty-eight and 72 h after renal ischemia, rats anesthetized with volatile anesthetic during renal ischemia and during 3 h of reperfusion had significantly lower plasma creatinine (fig. 1A) compared with rats anesthetized with pentobarbital or ketamine. Rats pretreated with volatile anesthetics for 1 h did not show improvements in renal function at 2 or 3 days after injury (fig. 1B).
The degree of renal necrotic injury is most severe in the distal proximal tubules or S3 segment located in the outer medullary area after ischemia–reperfusion injury. Therefore, we show this area in the figures. Compared with kidneys from sham-operated animals (fig. 2A), 45 min of renal ischemia during pentobarbital or ketamine anesthesia followed by 24 h of reperfusion resulted in significant renal injury demonstrated by severe tubular dilatation, tubular swelling and necrosis, medullary luminal congestion and hemorrhage, and development of proteinaceous casts (figs. 2B and C, respectively). Rats anesthetized with sevoflurane, halothane, or isoflurane during and after renal ischemia demonstrated markedly reduced histologic features of necrotic renal injury with preservation of near-normal renal histology after ischemia–reperfusion (figs. 2D–G). Desflurane also offered protection from renal tubular necrotic changes, but desflurane-mediated protection was less than that observed with other volatile agents. Rats pretreated with sevoflurane, isoflurane, halothane, or desflurane did not show improved renal morphology compared with pentobarbital-treated rats (data not shown). Quantitative assessment of renal tubular necrosis using the grading scores of Jablonski et al .20is shown in figure 3. Histologic grading at 24 h after 45 min of renal ischemia in pentobarbital- or ketamine-treated rats resulted in severe acute tubular necrosis. In contrast, rats treated with sevoflurane, isoflurane, or halothane showed significantly improved renal morphology compared with pentobarbital-treated rats. Desflurane-treated rats also showed improved renal necrosis scores compared with pentobarbital-treated rats; however, necrosis scores were significantly worse than with other volatile anesthetics. Rats pretreated with sevoflurane, isoflurane, halothane, or desflurane only before renal ischemia–reperfusion under pentobarbital anesthesia did not show improved quantitative assessment of renal tubular necrosis compared with pentobarbital-treated rats. Sham rats had a score of 0 ± 0 (n = 4).
We measured blood pressure in rats anesthetized with pentobarbital or with volatile anesthetics. After a transient (10- to 15-s) decrease (to approximately 110 mmHg), systolic blood pressure returned to normal (approximately 120–130 mmHg) during and after renal ischemia with pentobarbital or volatile anesthetics. As a representative example of volatile agents, sevoflurane produced a dose-dependent decrease in blood pressure at inhaled concentrations greater than 3% (fig. 4). However, at clinically relevant concentrations and the concentrations used in the current study (i.e ., approximately 2.2%, 1 MAC), the effects of sevoflurane on blood pressure were minimal as previously published.23–25Similar patterns were observed with other volatile anesthetics (data not shown).
Renal blood flow was estimated before and after ischemia–reperfusion injury. As a representative volatile agent, sevoflurane was compared to pentobarbital and ketamine anesthesia. There were significant reductions in renal blood flow 15 min into the reperfusion phase after 45 min of renal ischemia in rats receiving pentobarbital, sevoflurane, or ketamine anesthesia regimens. Before renal ischemia, renal blood flow values were 16.1 ± 1.7 ml · min−1· kg−1(n = 4), 22.6 ± 2.1 ml · min−1· kg−1(n = 3), and 22.1 ± 4.8 ml · min−1· kg−1(n = 4) in rats anesthetized with pentobarbital, sevoflurane, or ketamine, respectively. Sixty minutes after ischemia, during reperfusion, renal blood flow values were 7.6 ± 0.9 ml · min−1· kg−1(n = 4), 8.5 ± 0.3 ml · min−1· kg−1(n = 3), and 10.3 ± 1.3 ml · min−1· kg−1(n = 4) in rats anesthetized with pentobarbital, sevoflurane, or ketamine, respectively. Therefore, the renal protective effects of sevoflurane do not correlate with their effects on renal blood flow either before or after the induction of renal ischemia.
To determine whether volatile anesthetics protect renal function, at least in part, by reducing inflammation, we quantified renal inflammation in renal cortices after ischemia–reperfusion injury by five indices: (1) myeloperoxidase activity (marker of leukocyte infiltration), (2) amount of messenger RNA (mRNA) encoding markers of inflammation (TNF-α, ICAM-1, MCP-1, MIP-2, IL-8, and IP-10), (3) activation of nuclear translocation of proinflammatory transcription factor NF-κB, (4) expression of ICAM-1 protein, and (5) expression of TNF-α protein.
Renal cortices isolated from pentobarbital-treated rats subjected to 24 h of reperfusion after 45 min of renal ischemia showed increased myeloperoxidase activity compared with sham-operated control rats (fig. 5). Kidney extracts from sevoflurane-, isoflurane-, or halothane-treated rats had less myeloperoxidase activity at 24 h after injury than kidney extracts from pentobarbital-treated rats. The change in myeloperoxidase activity in kidneys from desflurane-treated rats did not reach statistical significance compared with rats anesthetized with pentobarbital and subjected to ischemia–reperfusion injury (P = 0.212).
Intercellular adhesion molecule 1 and TNF-α mRNA expression was significantly increased in renal cortices of rats undergoing renal ischemia–reperfusion during pentobarbital anesthesia as compared with rats undergoing sham operation during pentobarbital anesthesia. In renal cortices harvested from sevoflurane-, isoflurane-, or halothane-treated rats, this increase was significantly attenuated (representative gel images for sevoflurane shown in figs. 6A and B). Similarly, increases in mRNA encoding the chemokines MCP-1, MIP-2, IP-10, and IL-8 were also significantly attenuated in renal cortices from sevoflurane-, isoflurane-, and halothane-treated rats (figs. 6C–H). In marked contrast, desflurane anesthesia did not attenuate the increase in proinflammatory mRNAs seen with the other volatile agents (figs. 6C–H). In addition, volatile anesthetic pretreatment only before the onset of renal ischemia did not decrease the expression of proinflammatory mRNAs (data not shown), which is consistent with lack of renal protection afforded by volatile anesthetic pretreatment alone. Sham operation under volatile anesthetic versus pentobarbital anesthesia showed no difference in proinflammatory mRNA expression (data not shown).
Increased ICAM-1 protein expression was detected by immunoblotting in rat kidney cortex after ischemia–reperfusion injury during pentobarbital anesthesia. Volatile anesthetics reduced this ischemia-reperfusion–induced up-regulation of ICAM-1 protein expression. Figure 7is representative of five independent experiments showing sevoflurane as a representative anesthetic. Other volatile anesthetic also reduced the expression of ICAM-1 protein (data not shown).
We questioned whether the renal protective and antiinflammatory effects of volatile anesthetics are associated with the inhibition of the translocation of the transcription factor NF-κB. Binding of NF-κB to the nuclear fractions isolated from kidney cortex of rats that underwent sham operation, renal ischemia–reperfusion during pentobarbital anesthesia, or renal ischemia–reperfusion during sevoflurane anesthesia are shown in figure 8. Ischemia–reperfusion injury increases the binding of NF-κB, whereas sevoflurane treatment attenuated this increase in NF-κB binding.
We measured renal cortical TNF-α concentrations by enzyme-linked immunosorbent assay in sham-operated rats during pentobarbital anesthesia, pentobarbital-anesthetized rats subjected to IR, and rats subjected to ischemia–reperfusion under volatile anesthetics. Rats subjected to pentobarbital anesthesia and ischemia–reperfusion injury showed increased renal cortical concentrations of TNF-α (10.7 ± 1.1 ng TNF-α/mg protein, n = 3) compared with sham-operated rats (6.2 ± 0.4 ng/ml, n = 4). Rats anesthetized with sevoflurane (4.1 ± 0.4 ng/ml, n = 4), halothane (5.1 ± 0.3 ng/ml, n = 4), and isoflurane (4.6 ± 0.2 ng/ml, n = 4) and subjected to ischemia–reperfusion had significantly reduced concentrations of TNF-α. Rats injured with ischemia–reperfusion during desflurane anesthesia also had reduced renal TNF-α concentrations (7.4 ± 0.8 ng/ml, n = 3). However, this reduction in TNF-α with desflurane anesthesia was less than that observed with other volatile anesthetics.
We questioned whether volatile anesthetic opens K+ATPchannels and protects against renal ischemia–reperfusion injury. Rats were pretreated with 6 mg/kg intravenous glibenclamide, a K+ATPchannel antagonist, 30 min before sevoflurane anesthesia. Unlike their role in myocardial protection by ischemic preconditioning and volatile anesthetics, K+ATPchannels are not involved in volatile anesthetic mediated renal protection because renal protection persisted in the presence of glibenclamide (creatinine: 2.4 ± 0.4 mg/dl, n = 3).
The major findings of the current study are that clinically relevant concentrations (1 MAC) of volatile anesthetics given both during and after renal ischemia protect against renal ischemia–reperfusion injury in rats by dramatically reducing tubular necrosis typically observed after renal ischemia–reperfusion injury. In addition, volatile anesthetics showed differential protection from renal ischemia–reperfusion injury in that desflurane was less effective and showed less antiinflammatory effects compared with isoflurane, sevoflurane, or halothane. Moreover, two mechanistic differences were demonstrated in our renal studies compared with previous volatile anesthetic–mediated protection studies in the heart: (1) K+ATPchannels are not involved in volatile anesthetic–mediated protection from renal IR, and (2) volatile anesthetics must be present during renal ischemia–reperfusion to confer protection as opposed to the benefit realized in cardiac studies by volatile anesthetic pretreatment alone. Alterations in renal blood flow, although an important factor of determining renal function after ischemic injury, cannot account for the improved renal function and decreased necrosis associated with volatile anesthetic anesthesia.
Recent evidence shows protective effects of volatile anesthetic pretreatment against ischemia–reperfusion injury in the heart similar to the phenomenon of ischemic preconditioning.4,5The mechanisms of protection are unclear; however, several have been proposed, including mitochondrial and/or sarcolemmal K+ATPchannel activation,26–28protein kinase C activation,29,30and/or prevention of the increase and oscillation in intracellular Ca2+associated with ischemia–reperfusion.8We are aware that the mechanisms of renal protection may not parallel those of cardiac protection. For example, we have previously demonstrated that renal ischemic preconditioning and adenosine mediated renal protection signal via Giand protein kinase C but not through K+ATPchannels. In the current study, the protection afforded by volatile anesthetics against renal ischemia–reperfusion injury also did not involve K+ATPchannel activation because pretreatment with glibenclamide did not attenuate the protection afforded by sevoflurane. We have demonstrated previously that this dose of glibenclamide effectively blocks K+ATPchannels in rats.18
In this study, we demonstrated that pretreatment alone with volatile anesthetic did not protect renal function after ischemia–reperfusion injury, as has been previously demonstrated in heart.4,5Only rats treated with volatile anesthetics during both ischemia and reperfusion were protected against severe acute renal failure. Whether the protective effects of volatile anesthetics are induced during ischemia, reperfusion, or both was not elucidated in this study. Therefore, the mechanisms and timing of volatile anesthetic–mediated protection from ischemia–reperfusion injury in the kidney are fundamentally different protection in the heart.
In another study in isolated perfused rat liver, isoflurane, sevoflurane, and halothane reduced ischemia reperfusion injury when administered during the reperfusion phase; however, they did not reduce injury when administered only during ischemia.4–6
With acute renal failure, necrotic cell death results in further activation of inflammatory cascades, resulting in more severe secondary tissue damage.31,32Moreover, sublethal injury is amplified by the inflammatory and cytotoxic injury cascades activated during the reperfusion period.33–35The inflammatory component consists of the elaboration of cytokines (e.g ., TNF-α) and chemoattractive chemokines (e.g ., IL-8, MCP-1, MIP-2) and the expression of adhesion molecules (e.g ., ICAM-1, selectins) that cause leukocytes to accumulate in the vasa recta of the outer medulla.36–38Modulation of inflammatory cascades and ischemia–reperfusion injury by volatile anesthetics has been described in the heart,11,39lung,12,13,40and liver.6,41In the heart, volatile anesthetics protected against ischemia–reperfusion injury by reducing postischemic neutrophil adhesion in the coronary system,11most likely by preventing the up-regulation of CD11.10,39de Rossi et al .42showed that isoflurane attenuated the activation of three adhesion molecules involved in neutrophil activation (L-selectin, CD11a, and CD11b). Shayevitz et al .13showed that isoflurane and halothane significantly attenuated the inflammatory response (reduced myeloperoxidase activity and neutrophil infiltration) associated with multiorgan dysfunction syndrome. Mitsuhata et al .43described that volatile anesthetic exposure of human monocytes significantly attenuated the release of proinflammatory cytokines, IL-1β, and TNF-α. Therefore, these previous studies as well our current study support our hypothesis that volatile anesthetics may protect the kidney against ischemia–reperfusion injury by reducing necrotic and inflammatory renal cell death.
Our reverse-transcriptase polymerase chain reaction data show that mRNA for proinflammatory ICAM-1, TNF-α, and chemokines (MCP-1, MIP-2, IL-8, and IP-10) in renal cortices are up-regulated 24 h after ischemia–reperfusion injury. Increased free radicals generated during the reperfusion period increase the expression of cytokines (e.g ., TNF-α). The up-regulated cytokines as well as increased free radical production elicit the increased expression of chemokines (e.g ., MIP-1, MCP-1, IP-10, and IL-8). These chemokines are chemotactic to neutrophils, lymphocytes, and macrophages16,17and are major contributors to inflammatory injury after IR. Neutralization of several cytokines (MCP-1, MIP-2) attenuates injury after ischemia–reperfusion in the kidney.44,45We did not determine the source of cytokines and ICAM-1; however, proximal tubules, infiltrating leukocytes (neutrophils, macrophages, and lymphocytes) and endothelial cells can produce these proinflammatory agents. Attenuation of the expression of these cytokines, chemokines, and ICAM-1 by volatile anesthetics most likely contributed to the reduction in renal injury after IR.
After ischemia–reperfusion injury, proinflammatory signaling cascades are directed to the nucleus via proinflammatory transcription factors such as NF-κB. In the nucleus, these transcription factors bind to the specific DNA motif and regulate transcription of target genes including TNF-α, ICAM-1, and several key chemokines involved in inflammatory renal tubular damage (MCP-1, MIP-2, and IP-10).14,46,47Because the activation of NF-κB is an important event in the renal ischemia-reperfusion–induced inflammatory response, the effects of volatile anesthetics on NF-κB were investigated. Others have demonstrated that attenuation of NF-κB activation associated with ischemia–reperfusion injury reduces renal injury.48We have shown that NF-κB nuclear translocation is significantly reduced in rat kidney after volatile anesthetic treatment, and this may be a component of mechanism for renal protection with volatile anesthetics. These data are consistent with volatile anesthetics’ blunting of mRNA expression encoding proinflammatory cytokines, chemokines, and adhesion molecule and with the overall hypothesis that volatile anesthetic reduces proinflammatory responses after renal ischemia–reperfusion injury.
Our data show striking differences in renal protection between desflurane and sevoflurane. We have demonstrated that sevoflurane, isoflurane, halothane, and desflurane provide significant renal protection against ischemia–reperfusion injury. However, the degree of renal protection by desflurane was significantly less than that conferred by sevoflurane, isoflurane, or halothane. Differential physiologic effects of volatile anesthetics are well known. For example, halothane and isoflurane produce strikingly different inotropic and vasodilatory effects. In the experimental setting of rabbit and rat hearts, isoflurane, halothane, desflurane, and sevoflurane showed differences in attenuation of cellular injury and functional recovery suggesting different mechanisms of protection.49,50In addition, hepatic heme oxygenase-1 expression was differentially regulated by several volatile anesthetics; isoflurane and sevoflurane induced heme oxygenase-1 mRNA and protein, and desflurane did not induce this gene.51Whether these differences indicate differences in mechanism of renal protection or differences in cellular uptake of volatile anesthetics remains to be elucidated. Differences in lipid solubility might explain the differences in renal protective effects of volatile anesthetics.
In conclusion, we demonstrate significant but differential renal protection with volatile anesthetic treatment during and after renal ischemia. These studies suggest that the preferential use of certain volatile anesthetics during surgical procedures involving renal ischemia may impact postischemic renal function. The cellular mechanisms underlying this differential protective effect warrant further study.