Effects of Bupivacaine and Tetrodotoxin on Carrageenan-induced Hind Paw Inflammation in Rats (Part 1): Hyperalgesia, Edema, and Systemic Cytokines

All procedures were conducted in accordance with the Children’s Hospital Animal Care and Use Committee (Boston, Massachusetts). Young adult male Sprague-Dawley rats weighing 250–300 g were used. The animals were kept on a 12-h light–dark cycle with free access to food and water. The rats were handled repeatedly over at least 3 days before experiments to habituate them to the investigators and the testing paradigm.

Carrageenan was prepared fresh before each experiment (0.2 ml of 2% wt/vol solution of lambda carrageenan in saline (0.15 m, pH 7.4; Sigma Chemical Co., St. Louis, MO). Tetrodotoxin (50 μm) stock solutions were made by dissolving 1 mg tetrodotoxin (Sigma Chemical) in 10 ml of 20 mm sodium citrate buffer.14Bupivacaine, 0.5% wt/vol (5 g/l), with epinephrine was used as the LA (Sigma Chemical). Epinephrine from a commercial 1:1,000 (1 g/l) solution (American Regent Laboratories, Shirley, NY) was diluted to 55 μm (1:100,000) final concentration and prepared fresh for each set of injections.

Before nerve block injections, rats were anesthetized briefly with isoflurane (2–4% inspired concentration in 100% oxygen) by facemask. The block was initiated by introducing a 23-gauge needle posteromedially to the greater trochanter pointed in an anteromedial direction. When bone was touched, the needle was withdrawn 1 mm and the drug was injected. The final volume of injectate was 0.2 ml test solution. The left leg was always used for blocks. Previous experience in our laboratory with percutaneous sciatic blocks in rats has shown that precision is improved if the investigator always injects on the same (left) side, permitting injection with the dominant (right) hand while palpating with the left thumb.15,16 

To provide prolonged LA blockade while avoiding confounding proinflammatory and antiinflammatory effects induced by bupivacaine–dexamethasone microspheres themselves,17we used a model that ensured continuous blockade for a period of 15 h, by use of three modifications: (1) increasing the bupivacaine concentration to 0.5% and the volume of the injection to 0.2 ml, (2) addition of epinephrine, and (3) using a second injection 6 h later. Epinephrine was necessary to prolong block durations sufficiently and to limit systemic toxicity of tetrodotoxin.15,18All sciatic block injections and systemic (subcutaneous interscapular) injections were performed with solutions containing epinephrine at 55 μm (1:100,000) final concentration, so that each animal received the same total dose of epinephrine. Pilot experiments (data not shown) showed that rats receiving two sciatic block injections of either tetrodotoxin with epinephrine or bupivacaine with epinephrine in these doses at 6-h intervals maintained dense sensory and motor blockade (using von Frey filaments, thermal withdrawal latencies, and extensor postural thrust maneuvers) over the course of 15 h, as required for this paradigm.

Animals were assigned to 1 of 12 experimental groups (n = 9–12/group) as shown in table 1. Solutions for injections were prepared and then coded by coinvestigators so that the primary investigator (H.B.) was blinded to the contents of the injectates.

Each animal received three injections at time = 0: (1) a right (contralateral) or left (ipsilateral) subcutaneous hind paw injection (0.2 ml with 2% wt/vol carrageenan or saline), (2) a left sciatic block (0.2 ml with 0.5% bupivacaine, 50 μm tetrodotoxin, or 9 g/l saline), and (3) a systemic injection (subcutaneous interscapular with bupivacaine, tetrodotoxin, or saline). Tetrodotoxin or bupivacaine was only injected at one site (sciatic or back) in each animal. The animals received the same medications for a second left perisciatic injection and a second injection subcutaneously in the back 6 h later, to provide a prolonged effect of bupivacaine, tetrodotoxin, or saline (table 1and fig. 1).

All experiments were performed on a Bruker Biospec (Bruker Instruments, Inc., Billerica, MA) operating at 4.7 T with maximum gradient strength of 400 mT/m. A volume coil was used as a transmitter to ensure homogeneous excitation, while a surface coil of diameter 4 cm was used as a receiver to image both hind paws of anesthetized rats.

After matching, tuning, and acquisition of scout images, the hind paws were imaged in the coronal plane with a multislice T2-weighted rapid acquisition with relaxation enhancement imaging sequence (2,000-ms repetition time, 28-ms effective echo time, 1-mm slice thickness, and four signal averages). The matrix size was 128 × 128, and field of view was 4 × 4 cm², resulting in a spatial resolution of 312 × 312 μm². Injection of carrageenan solution into the hind paw resulted in observable increases of signal intensity within the muscle in the T2-weighted images. The edema signal was approximately as bright as the signal from bone marrow and subcutaneous fat. To perform quantitative assessments of the edema signal, the four or five slices containing edema from each study were thresholded such that only bright pixels containing edema, marrow, and fat remained (fig. 2), eliminating the darker signal from muscle and cortical bone. The total number of pixels containing edema, marrow, and subcutaneous fat was thus estimated for each edematous paw from the thresholded image. This number was divided by the total number of pixels in the image and multiplied by 100 to obtain an estimate of the percentage of tissue within each image containing edema, marrow, and subcutaneous fat. This approach allowed us to calculate and compare the relative percentage of edema signals between the different groups of rats, assuming similar volumes of marrow and subcutaneous fat and rough equivalence of paw volumes. We refer to the measurement performed in this way as a relative edema volume because the measurement also contains lipid signal.10 

All behavioral tests were performed by a single investigator (H.B.), who was blinded to the study groups. Behavioral tests were performed 15 h after initial injections, as outlined in figure 1. Thermal nociception was measured by a modified hot-plate test.19The time that a rat would leave its hind paw on a hot plate (model 39D Hot Plate Analgesia Meter; IITC Inc., Woodland Hills, CA) at 52°C was measured using a stopwatch (this time is called thermal latency). The paw was removed from the hot plate after 12 s by the investigator to avoid thermal injury and thermal hyperalgesia.19Failure to remove the hind paw after 12 s is regarded as dense thermal nocifensive blockade. This test was repeated three times on each hind paw for each rat.

The extensor postural thrust is a measure of motor strength in normal righting movements.20The rat was held above a digital balance and allowed to bear weight on one hind paw at a time. The maximum mass (in grams) that the rat could bear without its ankle touching the balance was measured.

The development of mechanical hypersensitivity after hind paw inflammation was assessed by the application of calibrated von Frey filaments. Animals were placed on a plastic mesh floor in individual plastic boxes and allowed to accommodate to their environment. Then, von Frey filaments were applied vertically to the plantar surface of both hind paws. Filaments were applied three times over 2 s. If no response was elicited, a larger diameter filament was applied in the same manner. The filaments were applied in increasing order until the brisk withdrawal or paw flinching was elicited, which was considered as a positive response. This withdrawal threshold was determined twice, with testing separated by 10 min, and the mean withdrawal threshold was used for data analysis. When no response is observed using the largest diameter, the animal is regarded to have a dense mechanical nocifensive sensory blockade.

As a secondary method to evaluate the edema, we used a technique previously described.1The paw circumference was measured by a thread, to the nearest mm, at the metatarsal level.

To reduce the confounding factors associated with the isolation of monocytes or neutrophils, such as adherence-induced activation or altered expression of cell-surface receptors, we induced cytokines production in whole blood cultures.21Blood was collected twice by tail venipuncture, immediately before drug injections (after induction of general anesthesia) and 15 h after drug injections, as outlined in figure 1. The collected blood was placed into a heparinized syringe (5 U heparin) from which whole blood cultures were performed. A blood sample (0.5 ml) was diluted 1:5 in RPMI-1640 medium (VWR, West Chester, PA) supplemented with antibiotics (penicillin–streptomycin; Sigma, St. Louis, MO). Five hundred–microliter aliquots of diluted blood were cultured in 20 four-well plates according to three different conditions: (1) without lipopolysaccharide or SAC (baseline), (2) with lipopolysaccharide (Escherichia coli  O111:B4, 10 μg/ml; Sigma), or (3) with SAC (100 μg/ml; Pansorbin Calbiochem, San Diego, CA) in a 5% CO²incubator for 24 h at 37°C. The supernatant was then harvested, centrifuged at 300g  at 4°C for 10 min, and stored at −70°C until assayed.

The amounts of TNF-α, interleukin 10 (IL-10), and IL-1β in the supernatants were measured with a commercial enzyme-linked immunosorbent assay kit (DuoSet; R&D systems, Minneapolis, MN) according to the manufacturer’s instructions. The assay detection limits were 30 pg/ml for TNF-α and IL-10, and 15 pg/ml for IL-1β.

At the end of each experiment, rats were killed with an overdose of pentobarbital (100 mg/kg administered intravenously; Nembutal, Abbott Laboratories, Chicago, IL).

Because the behavioral and the paw circumference data were not normally distributed, differences between groups were assessed using nonparametric tests (Kruskal-Wallis and Dunn tests). A stringent criterion (P < 0.01) was applied to protect against false positives (type I error) and account for multiple comparisons.22The results are expressed as median with 25th and 75th percentiles or as median with interval range for the von Frey data. Group differences in magnetic resonance imaging measurements of relative edema volumes (REVs) were assessed using a one-way analysis of variance with post hoc  analysis via  Fisher protected least significant difference test. The results are expressed as mean ± SD. A P  value below 0.01 was considered as the minimum level of statistical significance. The distribution of cytokine concentrations in each group was checked for normality using the Shapiro-Wilk test. Differences between groups were assessed using a one-way analysis of variance with post hoc  analysis via  Fisher protected least significant difference test. The results are expressed as mean ± SD. A P  value below 0.01 was considered as the minimum level of statistical significance.

Fifteen hours after the injection of carrageenan, a significant edema was observed in the hind paw carrageenan group (hind paw carrageenan, systemic saline, and contralateral saline) compared with the control group (hind paw saline, contralateral perisciatic saline, and systemic saline) as characterized by the REV measurements (table 2). In the groups receiving hind paw injection of saline along with either contralateral bupivacaine block or systemic bupivacaine, the REV was not significantly different from that of the control group receiving contralateral saline and systemic saline. In the groups receiving hind paw carrageenan plus bupivacaine as a contralateral block or as an ipsilateral block, the REV was significantly less than in the hind paw carrageenan group but still significantly greater than in the control group. In the group receiving hind paw carrageenan and systemic bupivacaine, the REV was not significantly different from that of the hind paw carrageenan group. For the non-hind paw–injected sides, there were no significant differences in the mean REVs in the different treatment groups. Figures 2A–Dshow coronal spin echo T2-weighted images before (black background) and after (blue background) application of a threshold criterion. Images show bilateral rat paws for the carrageenan group (A), the carrageenan and contralateral bupivacaine block group (B), the carrageenan and ipsilateral bupivacaine block group (C), and the carrageenan and systemic bupivacaine group (D).

Fifteen hours after the injection of carrageenan, a significantly increased paw circumference was observed in the hind paw carrageenan group compared with the control group (fig. 3). In the groups receiving hind paw carrageenan and an ipsilateral or contralateral sciatic block with either bupivacaine or tetrodotoxin, mean paw circumferences were significantly less than in the hind paw carrageenan group. Groups receiving hind paw carrageenan plus systemic tetrodotoxin or bupivacaine had mean paw circumferences that were not significantly different from those of the hind paw carrageenan group (fig. 3).

No significant differences were observed between the groups (data not shown).

Fifteen hours after injection of carrageenan, a significant decrease in heat withdrawal latency was observed in the carrageenan group compared with the control group, indicating thermal hyperalgesia in the carrageenan group (fig. 4). Contralateral (left) sciatic injection of bupivacaine or tetrodotoxin had no effect on right hind paw withdrawal latencies in groups receiving right hind paw saline. In groups receiving right hind paw carrageenan and a contralateral block with either bupivacaine or tetrodotoxin, withdrawal latencies were significantly greater than in the carrageenan group, indicating partial prevention of carrageenan-induced thermal hyperalgesia. Systemic bupivacaine and tetrodotoxin were ineffective in prevention of carrageenan-induced thermal hyperalgesia. In groups receiving left hind paw carrageenan and an ipsilateral sciatic block with either bupivacaine or tetrodotoxin, withdrawal latencies were at the cutoff value of 12 s, indicating dense thermal nocifensive blockade even under conditions of hind paw inflammation (fig. 4).

In the groups receiving a sciatic block with either bupivacaine or tetrodotoxin, the animals showed dense thermal nocifensive blockade; otherwise, there were no differences between groups in mean thermal latencies (data not shown). Therefore, carrageenan-induced inflammation produced ipsilateral, but not contralateral, hind paw thermal hyperalgesia.

Fifteen hours after the injection of carrageenan, a significant decrease in the withdrawal threshold was observed in the carrageenan group compared with the control group, indicating mechanical hyperalgesia (fig. 5).

Contralateral (left) sciatic injection of bupivacaine or tetrodotoxin had no effect on the right hind paw withdrawal thresholds in groups receiving right hind paw saline. In the groups receiving right hind paw carrageenan and a contralateral sciatic block with either bupivacaine or tetrodotoxin, withdrawal thresholds were significantly greater than in the carrageenan group, indicating partial prevention of carrageenan-induced mechanical hyperalgesia. Systemic bupivacaine and tetrodotoxin were ineffective in prevention of carrageenan-induced mechanical hyperalgesia. In the groups receiving left hind paw carrageenan and an ipsilateral block with either bupivacaine or tetrodotoxin, mean withdrawal thresholds were 60 g or greater, indicating dense mechanical nocifensive blockade, even under conditions of hind paw inflammation (fig. 5).

For groups receiving right-sided carrageenan or saline injections, animals receiving left-sided sciatic blocks with either bupivacaine or tetrodotoxin all showed dense mechanical nocifensive blockade. For the other groups, there were no differences in mean withdrawal thresholds (data not shown). Therefore, carrageenan-induced inflammation produced ipsilateral, but not contralateral, hind paw mechanical hyperalgesia.

Fifteen hours after the injection of carrageenan, a significant decrease in the weight an animal could bear on the injected (right) paw was observed in the carrageenan group compared with the control group (fig. 6). Sciatic injection of either bupivacaine or tetrodotoxin had no effect on contralateral (right) weight bearing compared with controls. In the groups receiving carrageenan and a contralateral block with either bupivacaine or tetrodotoxin, the weight an animal could bear in the right paw was significantly greater than in the carrageenan group but still significantly lower than in the control group. Ipsilateral sciatic blockade with either tetrodotoxin or bupivacaine in animals receiving left hind paw injections produced marked impairment of weight bearing (fig. 6).

For all treatment conditions, weight bearing was markedly impaired in the limbs that had received a sciatic block with either bupivacaine or tetrodotoxin, and unimpaired in all other groups (data not shown).

Concentrations of TNF-α, IL-1β, and IL-10 in cultured blood in the absence of either lipopolysaccharide or SAC stimulation were always low in all of the study groups obtained by venipuncture at H0, indicating that incubation in the culture plates per se  did not significantly stimulate cytokine production in cultured cells from these groups.

Lipopolysaccharide or SAC stimulation of blood cultures obtained by venipuncture at H0 produced concentrations of each of the three cytokines that were statistically indistinguishable among all treatment conditions (hind paw carrageenan or saline; tetrodotoxin, bupivacaine, or saline at all three sites; data not shown). Concentrations of each of the three cytokines under these lipopolysaccharide- or SAC-stimulated incubation conditions were also statistically indistinguishable from concentrations of the same cytokines evoked in blood cultures in the control groups obtained by venipuncture at H15 (see previous section and figs. 7–9; data not shown). The observation of lower amounts of evoked production of cytokines in the carrageenan-treated H0 treatment groups compared with the carrageenan-treated H15 groups is consistent with the view that the additional increases in stimulated cytokine production in the H15 groups reflect a systemic leukocyte activation response to carrageenan that evolves over time.

Concentrations of TNF-α, IL-1β, and IL-10 in cultured blood obtained at H15 in the absence of either lipopolysaccharide or SAC stimulation were low in all groups, indicating that the culture plates per se  did not significantly stimulate cytokine production in cultured cells from these groups.

Twenty-four hours after stimulation with lipopolysaccharide (fig. 7A) or SAC (fig. 7B), TNF-α production was enhanced in blood cultures obtained 15 h after hind paw saline (control group). A significant further increase in TNF-α production after lipopolysaccharide or SAC stimulation was observed in blood cultures obtained 15 h after hind paw carrageenan injection. In the groups receiving hind paw saline and either bupivacaine or tetrodotoxin by all three routes, TNF-α production was not significantly different from the production observed in the control group. In the groups receiving hind paw carrageenan and bupivacaine by all three routes, lipopolysaccharide-stimulated TNF-α production was not significantly different from the production observed in the control group, and it was significantly less than in the carrageenan group, showing that bupivacaine by all three routes markedly inhibited the increased production of TNF-α produced by hind paw carrageenan. After SAC stimulation, in the groups receiving carrageenan and bupivacaine, TNF-α production was significantly higher than in the control group and significantly lower than in the carrageenan group, showing that bupivacaine by all three routes had a statistically significant, but incomplete, suppressive effect on SAC-stimulated TNF-α production (fig. 7B). Tetrodotoxin by all three routes did not have any effect on the carrageenan-induced TNF-α production after lipopolysaccharide or SAC stimulation.

Twenty-four hours after stimulation with lipopolysaccharide (fig. 8A) or SAC (fig. 8B), IL-1β production was enhanced in the hind paw saline (control) group. A significant further increase in IL-1β production after lipopolysaccharide or SAC stimulation was observed 15 h after hind paw carrageenan injection. In the groups receiving hind paw saline and bupivacaine or tetrodotoxin by all three routes, lipopolysaccharide- or SAC-stimulated IL-1β production was not significantly different from the production observed in the control group. In the groups receiving hind paw carrageenan and bupivacaine by all three routes, lipopolysaccharide-stimulated IL-1β production was not significantly different from the production observed in the control group and was significantly lower than in the carrageenan group, showing that bupivacaine by all three routes markedly inhibits carrageenan-induced increased production of IL-1β after lipopolysaccharide stimulation (fig. 8A). In blood cultures from these same treatment groups receiving hind paw carrageenan, SAC-stimulated IL-1β production was significantly higher than in the control group, and there were no significant differences between all of the groups receiving hind paw carrageenan (fig. 8B). Tetrodotoxin by all three routes did not have any effect on the carrageenan-induced increases in IL-1β production after lipopolysaccharide (fig. 8A) or SAC (fig. 8B) stimulation.

Twenty-four hours after stimulation with lipopolysaccharide (fig. 9A) or SAC (fig. 9B), IL-10 production was enhanced (control group). Lipopolysaccharide or SAC induced further increases in IL-10 production in all of the groups receiving hind paw carrageenan as compared with all of the groups receiving hind paw saline. No differences were observed between the groups receiving carrageenan. Therefore, tetrodotoxin and bupivacaine, by any route, did not prevent carrageenan-induced increases in IL-10 production after either lipopolysaccharide or SAC stimulation.

In the current study, both contralateral and ipsilateral sciatic blockade using bupivacaine significantly attenuated carrageenan-induced unilateral hind paw inflammatory edema. Contralateral bupivacaine sciatic blocks significantly attenuated carrageenan-induced mechanical and thermal hyperalgesia, whereas ipsilateral sciatic blocks produced dense sensory and motor blockade even in the presence of this inflammatory stimulus. Systemic administration of bupivacaine did not affect either hyperalgesia or edema. In view of the multiple molecular targets of amino-amide LAs such as bupivacaine, the same experiments were also performed with tetrodotoxin, which seems to act much more selectively on sodium channels. In the current study, bupivacaine and tetrodotoxin showed similar impact in suppressing local hind paw edema when administered by ipsilateral or contralateral sciatic block, similar suppression of hyperalgesia when administered by contralateral sciatic block, dense sensory and motor block when administered by ipsilateral block (even in the presence of inflammation), and similar lack of impact on edema and hyperalgesia when administered by remote systemic injection. These results suggest that ipsilateral or contralateral sciatic injection of bupivacaine or tetrodotoxin, acting by regional rather than systemic mechanisms, inhibit carrageenan-induced local hind paw edema and hyperalgesia. In contrast, bupivacaine and tetrodotoxin showed markedly different effects on a systemic inflammatory marker, namely lipopolysaccharide-stimulated release of the cytokines TNF-α and IL-1β in cultures of circulating blood cells. Bupivacaine was effective in suppressing these systemic inflammatory markers by all three routes, whereas tetrodotoxin was ineffective by all three routes. Table 3summarizes these differences in actions of bupivacaine and tetrodotoxin on local and systemic responses to carrageenan-induced inflammation.

Previous studies of effects of LAs on carrageenan-induced hyperalgesia, edema, and inflammation have yielded some conflicting results.1–3,6,7,9Some of these differences may result from the different sites of LA delivery, different methods used to achieve prolonged effect, and the confounding proinflammatory and antiinflammatory effects of microspheres and coincorporated dexamethasone.17As detailed in the Materials and Methods, to avoid some of these confounding factors, we used three modifications to provide continuous blockade for the required 15-h period for both bupivacaine and tetrodotoxin: (1) increasing the bupivacaine concentration to 0.5% and the volume of the injection to 0.2 ml, (2) addition of epinephrine, and (3) using a second injection 6 h later.

In accordance with previous studies,9we observed that carrageenan induced ipsilateral edema and hyperalgesia without inducing either edema or hyperalgesia in the contralateral hind paw. Conversely, one article3had shown a bilateral decrease in hind paw withdrawal thresholds after unilateral carrageenan injection. In the current study, bupivacaine or tetrodotoxin via  a sciatic block did not influence the contralateral hind paw behavioral measures, and systemic bupivacaine or tetrodotoxin had no effect on hind paw edema or behavioral measures in noninjected paws.

In accordance with previous studies,1,2the local hyperalgesia and edema induced by carrageenan was partly inhibited by a bupivacaine block. Gentili et al.  2described a significant decrease of carrageenan-induced edema in rats related to a prolonged (> 6 h) peripheral nerve block. This antiinflammatory effect was associated with a persistent analgesic effect. Of interest, in accordance with the results of Bileviciute-Ljungar and Lundeberg,3an injection in the leg contralateral to the carrageenan injection had the same effect as an ipsilateral injection. Moreover, systemic bupivacaine did not affect the pain-related behavior and the edema. The impact of contralateral LA administration in our study and the study of Bileviciute-Ljungar and Lundeberg was remarkable and not readily explained.

In addition to local, unilateral inflammation, the carrageenan model evokes a systemic inflammatory response, as evidenced by increased TNF-α, IL-1β, and IL-10 production in cultures of circulating leukocytes after lipopolysaccharide or SAC stimulation.23LAs have been shown in previous studies using carrageenan-induced inflammation to influence systemic as well as local inflammatory responses6,7,24,25In our previous study, contralateral intramuscular administration of bupivacaine microspheres in mice inhibited evoked cytokine (TNF-α and IL-1β) release in circulating leukocytes.6In the current study, increased production of TNF-α and IL-1β in cultures of circulating blood cells after lipopolysaccharide stimulation in rats receiving hind paw carrageenan injections was completely prevented by bupivacaine treatment, via  ipsilateral or contralateral sciatic block or systemically. In the case of SAC-stimulated blood cultures, evoked increases in TNF-α production were partially prevented by bupivacaine by any of these three routes, and evoked increases in IL-1β production were not prevented by bupivacaine by any route. The administration of bupivacaine had no effect on the carrageenan-induced IL-10 production after either lipopolysaccharide or SAC stimulation. In the absence of carrageenan-induced hind paw inflammation, bupivacaine by any route did not influence systemic cytokine production. The actions of lipopolysaccharide and SAC on mononuclear cells leading to TNF-α and IL-1β release involve different Toll-like receptors and different signaling pathways.26Differential actions of bupivacaine on different Toll-like receptors and signaling pathways may be a possible explanation of our observations of differential effects of bupivacaine on the inflammatory profiles elicited by lipopolysaccharide or SAC in the animals receiving carrageenan.

Tetrodotoxin had no effect on systemic cytokine production when administered by any of the three routes. These differences in actions of bupivacaine and tetrodotoxin on these markers of systemic inflammation were unexpected, because bupivacaine and tetrodotoxin had similar effects on amelioration of hind paw carrageenan-induced local inflammatory edema and hyperalgesia. Taken together, these results would suggest that LAs may influence local and systemic inflammatory responses by different mechanisms.

The local tissue neurogenic inflammatory response to injury is generally attributed to an efferent function of primary afferent terminals mediated through axon reflexes. It can also be initiated in the spinal cord by primary afferent depolarization large enough to trigger dorsal root reflexes. Activation of dorsal root reflexes was shown to play a role in the development of neurogenic cutaneous inflammation.27The different local and systemic consequences of ipsilateral, contralateral, and systemically applied bupivacaine and tetrodotoxin could possibly relate to an effect on dorsal root reflex activation. Duarte et al.  28recently reported additional evidence for the effectiveness of bupivacaine for suppressing postoperative pain through a mechanism other than a direct blockade of afferent discharges at the incision site. In this study, allodynia, particularly in the secondary areas, was selectively suppressed for at least a week by preincisional bupivacaine given subcutaneously at the incision site and elsewhere. Differential effects of drug and site of administration on expression of cytokines and mitogen-activated protein kinases in dorsal root ganglia and in the spinal cord could also be involved.29,30A unilateral nerve injury can produce bilateral loss of distal innervation,31and ipsilateral and mirror-image inflammatory neuropathic pain can be created acutely and chronically through glial and proinflammatory cytokine actions.32Additional studies to investigate the effects of LAs on spinal mechanisms involved in carrageenan-induced hind paw inflammation will be reported separately.

The clinical implications of local, systemic, and spinal inflammatory effects of LAs require further study. In a recent study, preemptive epidural analgesia (mixture of bupivacaine and fentanyl) reduced postoperative pain and production of proinflammatory cytokines33but did not alter production of IL-10. IL-10 inhibits lipopolysaccharide-induced production of proinflammatory cytokines by mononuclear cells under in vitro  and in vivo  conditions.34Production of IL-10 has therefore been considered to a part of a host-protective mechanism during excessive inflammatory processes. In our experiments, both challenges (lipopolysaccharide and SAC) stimulated the production of IL-10 by the cultured cells, and carrageenan enhanced this production, but neither bupivacaine nor tetrodotoxin by any of the three routes influenced evoked production of IL-10.

A potential trivial mechanism for the effects of contralateral sciatic blocks would be if drugs spread proximally from injection sites in the buttocks back into the epidural space. Two observations argue against this possibility. First, previous studies using block injections labeled with fluorescent dyes showed very little rostral spread, never close to the spine.35Second, sciatic blocks using this technique, either in the current study or in previous studies using larger volumes,15have never demonstrated significant degrees of contralateral sensory or motor blockade by behavioral measures.

Another potentially confounding feature of this model is that coinjection of epinephrine is required to prolong block durations sufficiently and to limit systemic toxicity of tetrodotoxin.18,35Pilot experiments showed that rats receiving two sciatic block injections of either tetrodotoxin with epinephrine or bupivacaine with epinephrine at 6-h intervals maintained dense sensory and motor blockade (using von Frey filaments, thermal withdrawal latencies, and extensor postural thrust maneuvers) over the course of 15 h, as required for this paradigm. Because epinephrine can decrease proinflammatory cytokine expression,36we cannot entirely exclude a possible effect of epinephrine on cytokine production or a cooperative interaction between epinephrine and bupivacaine. However, tetrodotoxin injected with epinephrine showed no effect on cytokine production. Moreover, control (saline-injected) groups received the same doses of epinephrine as the bupivacaine and tetrodotoxin groups, to account for both local and systemic actions of epinephrine.

In conclusion, bupivacaine and tetrodotoxin via  a contralateral or ipsilateral sciatic block attenuate local inflammatory edema and hyperalgesia induced by carrageenan in rats. Bupivacaine, by all three routes, inhibits carrageenan-induced stimulated activation of systemic cytokines, whereas tetrodotoxin by all three routes is ineffective. Bupivacaine may possibly exert systemic antiinflammatory effects via  actions on tetrodotoxin-resistant sodium channels or by a variety of mechanisms unrelated to either sodium channels or conduction blockade. Local, systemic, and spinal inflammatory mechanisms may be influenced by LAs through multiple different mechanisms. The therapeutic implications of effects of LAs on local, systemic, and spinal inflammatory responses merit further study.

Jennifer Jaskiewicz and Michelle Stein (Research Assistants, Department of Anesthesiology, Perioperative and Pain Medicine, Children’s Hospital, Boston, Massachusetts) provided excellent technical assistance in preparation of figures. The authors would like to thank Dan Benhamou and Jean-Xavier Mazoit (Professors, Hopital de Bicetre, Paris, France) for their encouragement and support of this work, which is a portion of Dr. Beloeil’s Ph.D. thesis at the University of Paris.