Effect of intravenous lidocaine administration on laminar inflammation in the black walnut extract model of laminitis



    Search for more papers by this author
    • *

      Author to whom correspondence should be addressed.

  • Y. J. LIN,

    1. Department of Clinical Sciences, The Ohio State University, College of Veterinary Medicine, Columbus, Ohio 43210; Department of Pharmacal Sciences, Auburn University, Auburn, Alabama 36849;
    Search for more papers by this author
  • J. P. LOFTUS,

    1. Department of Veterinary and Animal Sciences, University of Massachusetts, Amherst, Massachusetts 01003; and
    Search for more papers by this author

    1. Department of Clinical Sciences, The Ohio State University, College of Veterinary Medicine, Columbus, Ohio 43210; Department of Pharmacal Sciences, Auburn University, Auburn, Alabama 36849;
    Search for more papers by this author
  • J. F. PERONI,

    1. Department of Large Animal Medicine, University of Georgia, Athens, Georgia 30602, USA.
    Search for more papers by this author
  • J. A. E. HUBBELL,

    1. Department of Clinical Sciences, The Ohio State University, College of Veterinary Medicine, Columbus, Ohio 43210; Department of Pharmacal Sciences, Auburn University, Auburn, Alabama 36849;
    Search for more papers by this author
  • W. R. RAVIS,

    1. Department of Clinical Sciences, The Ohio State University, College of Veterinary Medicine, Columbus, Ohio 43210; Department of Pharmacal Sciences, Auburn University, Auburn, Alabama 36849;
    Search for more papers by this author

    1. Department of Clinical Sciences, The Ohio State University, College of Veterinary Medicine, Columbus, Ohio 43210; Department of Pharmacal Sciences, Auburn University, Auburn, Alabama 36849;
    Search for more papers by this author


Reasons for performing study: Laminitis is a serious complication of horses suffering from sepsis/endotoxaemia-related events. Laminitis in horses and organ injury in human sepsis are both reported to involve inflammatory injury to the laminae/organs including early activation of endothelium and leucocytes leading to emigration of neutrophils into the tissue interstitium. In the black walnut extract (BWE) model, systemic inflammatory events coincide with marked increase in laminar mRNA concentrations of inflammatory genes including proinflammatory cytokines (i.e. IL-1β, IL-6), COX-2, chemokines (i.e. IL-8) and endothelial adhesion molecules (i.e. ICAM-1 and E-selectin). In models of human sepsis, i.v. lidocaine has been reported to decrease leucocyte and endothelial activation, and the expression of proinflammatory cytokines and chemokines.

Objectives: To evaluate the effect of i.v. lidocaine therapy on the inflammatory processes documented to occur in the BWE model of laminitis.

Methods: Twelve horses were administered BWE and treated immediately with either lidocaine (1.3 mg/kg bwt bolus, followed by 0.05 mg/kg bwt/min CRI, n = 6) or saline (n = 6) for 10 h. At 10 h post BWE administration, laminar samples were obtained under general anaesthesia for assessment of proinflammatory gene expression (using RT-qPCR) and leucocyte emigration (via CD13 immunohistochemistry). At 0, 3 and 10 h post BWE administration, skin samples were obtained for assessment of leucocyte emigration (via calprotectin immunohistochemistry).

Results: No significant differences between groups were noted for inflammatory gene mRNA concentrations (IL-1β, IL-6, IL-8, COX-2) or for number of leucocytes present within the laminar interstitium or skin dermis. Increased (P<0.05) laminar E-selectin mRNA concentrations were present in the LD group (vs. SAL group).

Conclusions: Continuous administration of i.v. lidocaine does not inhibit inflammatory events in either the laminae or skin in the horse administered black walnut extract.

Potential relevance: This work questions the use of continuous i.v. administration of lidocaine as an effective anti-inflammatory therapy for systemic inflammation.


Lidocaine hydrochloride has been used extensively in human and veterinary medicine to provide local and regional anaesthesia, and to treat and prevent ventricular arrhythmia (Mama and Steffey 2001; Plumb 2005). Systemically, its use in veterinary medicine, particularly in equine medicine, has been as a treatment for post operative ileus (Brianceau et al. 2002; Cohen et al. 2004; Malone et al. 2006). Lidocaine has also been reported to have inhibitory effects on a broad spectrum of systemic inflammatory processes (Hollmann and Durieux 2000). It has been shown to inhibit neutrophil adhesion and migration (Schmid et al. 1996), reduce free radical production by neutrophils, and inhibit the phagocytic activity of leucocytes (Siminiak and Wysocki 1992; Hyvonen and Kowolik 1998). Lidocaine reportedly decreases gene expression of various cytokines (i.e. IL-1β, IL-6, IL-8) in epithelial cells, endothelial cells and neutrophils (Sinclair et al. 1993; Lahav et al. 2002; Lan et al. 2005) and attenuates cytokine-induced endothelial injury (de Klaver et al. 2003) and endothelial cell expression of adhesion molecules important in leucocyte adhesion and extravasation (i.e. ICAM-1) (Lan et al. 2005).

Laminitis is a serious disease process in which laminar injury occurs as a remote organ injury in animals suffering from a wide array of septic conditions. Acutely, laminitis occurs secondary to numerous disease states affecting of the gastrointestinal, respiratory, reproductive, endocrine or musculoskeletal systems (Hood 1999); and can be induced experimentally with carbohydrate overload or extract from black walnut heartwood (BWE) (Garner et al. 1975; Galey et al. 1991). Using the BWE model, researchers have detailed both systemic and local inflammatory processes in the laminae in the early developmental stages of the disease process, with many similarities to inflammatory events leading to organ injury in human sepsis.

These BWE-induced inflammatory events include activation of circulating leucocytes (Hurley et al. 2006), activation of adhesion molecule expression by the laminar endothelium (Loftus et al. 2007) and upregulation of laminar gene expression of proinflammatory mediators (Fontaine et al. 2001; Belknap et al. 2007; Loftus et al. 2007). This inflammatory process in the BWE model is characterised by increased laminar mRNA concentrations of proinflammatory cytokines/chemokines (IL-1β, IL-6 and IL-8), increased laminar mRNA and protein concentrations of cyclooxygenase-2 (COX-2), and an influx of inflammatory leucocytes (neutrophils and monocytes) into the dermal laminae at both the developmental stage of laminitis and at the acute onset of clinical signs (Waguespack et al. 2004a,b; Black et al. 2006; Blikslager et al. 2006; Belknap et al. 2007; Loftus et al. 2007; Faleiros et al. 2009).

Due to the reported inhibitory effects of systemic lidocaine on the majority of inflammatory events reported to occur in the laminae in the BWE laminitis model, and its proposed use as an anti-inflammatory therapy in human inflammatory conditions including systemic inflammation in sepsis (Siminiak and Wysocki 1992; Sinclair et al. 1993; Schmid et al. 1996, Hyvonen and Kowolik 1998; Hollmann and Durieux 2000; Lahav et al. 2002; de Klaver et al. 2003; Lan et al. 2005), the effect was evaluated of a constant rate infusion (CRI) of lidocaine on multiple inflammatory processes that occur in the BWE model of laminitis. It was hypothesised that a CRI of lidocaine would markedly decrease laminar inflammatory events including endothelial activation, leucocyte emigration into the laminar tissue, and proinflammatory cytokine and chemokine mRNA concentrations.

Materials and methods


Twelve healthy Standardbred horses (age 3–12 years) with no apparent forelimb foot abnormalities or forelimb lameness (as determined by lameness evaluations and hoof tester application) were obtained from an outbred population. The horses were quarantined and kept together for 2 weeks at pasture at a private boarding facility. They were then moved to the Ohio State University Finley Farms facility and kept in quarantine for 2 weeks. While quarantined, physical examinations including assessment of rectal temperature, heart rate, respiratory rate, abdominal sounds and digital pulses were performed daily. The horses were moved subsequently to the Galbreath Equine Center at the Ohio State University for completion of the study.

Experimental protocol

All animal protocols were approved by the Institutional Animal Care and Use Committees of the Ohio State University. The health status of the horses was evaluated by physical and lameness examination on the day of experiment to ensure no concurrent illness that would affect peripheral white blood cell count and to ensure no forelimb lameness. An indwelling catheter was placed into each jugular vein, one for blood sample collection and the other for constant rate infusion of lidocaine or saline. A blood sample was obtained for haematology and plasma was used for determining baseline lidocaine concentrations. Skin samples were obtained with a 6 mm diameter skin biopsy punch from the cranial aspect of the right neck (0 h time point) and caudal aspect of the right neck (3 h time point) following local skin anaesthesia with 1 ml of mepivacaine per site. The BWE was made by soaking 2 g black walnut heartwood shavings/kg bwt in 6 l deionised water as previously described (Belknap et al. 2007). Six litres of BWE were administered via nasogastric tube. Horses were assigned randomly to one of 2 treatment groups, with 6 horses in each group: 1) lidocaine 1.3 mg/kg bwt loading dose administered over 15 min, followed by 0.05 mg/kg bwt/min constant rate infusion; or 2) 0.9% saline administered as a bolus and constant rate infusion at calculations based upon lidocaine administration. The administrators of lidocaine/saline were blinded to the contents of each therapy, and therefore calculated each administration as if it were lidocaine.

Physical examinations were performed by observers blinded to treatment and included assessment of vital parameters, abdominal sounds, digital pulses, attitude, mentation, signs of colic (lying down and rolling, flank watching, Flehmen response), weight shifting. Application of hoof testers was performed at 0, 1.5, 3, 6, 8, 9 and 10 h post BWE administration. Blood was obtained for haematology at 0, 3 and 10 h post BWE administration. Skin samples were obtained as described above at 0, 3 and 10 h post BWE administration and were snap frozen immediately in liquid nitrogen in addition to formalin fixation. At 10 h post BWE administration, horses were sedated with xylazine (1.1 mg/kg bwt, i.v.) and anaesthetised using diazepam (0.1 mg/kg bwt, i.v.) and ketamine (2.2 mg/kg bwt, i.v.). The horses were intubated orotracheally and a deep surgical plane of anaesthesia maintained with isofluorane during sample collection (see below). Intravenous constant rate infusions of lidocaine or saline were continued throughout the anaesthetic procedure until euthanasia with pentobarbital sodium containing phenytoin sodium (Beuthanasia-D)1 (20 mg/kg bwt, i.v.).

Sample collection

Each distal limb was removed rapidly by disarticulation of the metacarpo- and metatarsophalangeal joints after placement of a tourniquet, and 1.5 cm thick sagittal sections of the digit were immediately cut with a band saw. The laminae were dissected rapidly from the hoof and third phalanx, and some sections were snap frozen immediately in liquid nitrogen, while the others were placed in 10% formalin. The processing time from disarticulation to placement of samples in formalin or liquid nitrogen was approximately 5 min. All horses were subjected to euthanasia with pentobarbital sodium (1 ml/4.54 kg bwt) while under anaesthesia at the end of the protocol. Samples fixed in formalin were transferred to 70% ethanol 24 h after collection and stored at 4°C until paraffin-embedded. Flash frozen samples were kept in liquid nitrogen and transferred to -80°C approximately 30 min after collection. All skin samples were frozen in liquid nitrogen or fixed in formalin in an identical manner to the laminar samples.


Formalin-fixed laminar and skin samples were paraffin-embedded and sectioned. Sample slides were created with multiple paraffinembedded sections per slide, and routine immunohistochemical methods were performed as previously described (Lunn et al. 1998; Black et al. 2006; Faleiros et al. 2009). Briefly, slides were deparaffinised and antigen retrieval performed using citrate buffer. An anti-equine CD13 monoclonal antibody (courtesy of Dr D. Paul Lunn, Department of Clinical Sciences, College of Veterinary Medicine and Biomedical Sciences, Colorado State University, Fort Collins, Colorado, USA) was used as a marker to assess the presence of leuococytes (neutrophils and monocytes) in the laminae in order to compare results to those previously reported in the BWE model (Black et al. 2006), while a mouse anti-human calprotectin monoclonal antibody (MAC387)2 was used as a marker to assess the presence of leucocytes (neutrophils and monocytes) in the skin (calprotectin staining more effectively detected skin leucocytes compared to CD13 in a pilot experiment).

Localisation of CD13 and calprotectin antigens was performed via routine IHC protocols as previously described (Black et al. 2006; Faleiros et al. 2009), using a commercial immunoperoxidase kit (Vectastain Elite ABC Kit)3 and 3′3 diaminobenzidine hydrochloride (DAB- Peroxidase Substrate Kit)3. Slides were assessed by an investigator blinded to the horse and treatment group from which the sections were obtained. Laminar CD13-positive cell counts were performed by centring the 40x objective on a laminar vessel in the primary dermal lamina, and all positively stained cells within the field were counted. Thirty (10 fields per section × 3 sections per slide × 1 slide per horse) 40x fields were counted from each horse and averaged yielding one CD13-positive cell count per horse. Skin calprotectin-positive cell counts were performed by taking 0.54 × 0.46 mm sized photocaptures (Image Scope)4 of each sample. Twenty photocaptures were obtained for each sample, and counts were performed electronically by Image J5. The blinded observer ensured that cells highlighted by the program were appropriately labelled leucocytes. The counts were then averaged together to yield one count per time point per horse.

Plasma lidocaine concentrations

Venous blood samples were obtained for measurement of plasma lidocaine concentrations at 0, 3 and 10 h post BWE administration. Plasma lidocaine concentrations were determined by an analytical method modified from a procedure previously described by Rofael and Abdel-Rahman (2002). An internal standard (ketamine) and borate buffer (pH 9.0) were added to plasma samples followed by extraction with a mixture of isopropanol-chloroform (1:9, v/v). The mixture was vortexed for 1 min and then centrifuged at 1000 g for 10 min at 4°C. The organic layer was aspirated and dried under a gentle stream of air. The residue was reconstituted with mobile phase and analysed with a Waters HPLC system6.

The aqueous portion (65%) of mobile phase consisted of 100 mmol/l monobasic phosphate with 30 mmol/l triethylamine dissolved in distilled water. The organic portion (35%) of the mobile phase consisted of 60% acetonitrile and 40% of methanol (v/v). The HPLC system was equipped with a Phenomenex precolumn (SecurityGuard)7 and a C-18 reversed phase column (Luna 5 µm, C18, 100A, 100 × 4.6 mm)7. The mobile phase flow rate was 1.5 ml/min and UV detector (Jasco UV 2075 plus)8 was performed at 210 nm.

The resolution times for ketamine and lidocaine were 9 and 12.5 min, respectively. A linear equation was obtained for plasma concentrations of 40–2808 ng/ml (correlation coefficient: 0.9994). The extraction efficiency for lidocaine from plasma was 97%. Both intra- and interday assay variations were <10%. The limit of detection in plasma was 10 ng/ml and the limit of quantification was 40 ng/ml.

RNA isolation and cDNA synthesis

Total RNA was extracted from the forelimb laminae of each horse (Absolutely RNA Miniprep)9. PolyA mRNA was then isolated (mRNA Extraction Kit)10 and used to make cDNA via reverse transcription (Retroscript)11. The cDNA was used to perform a real-time quantitative polymerase chain reaction (RT-qPCR) to determine mRNA concentration of different genes of interest.

Real time-quantitative PCR (RT-qPCR) procedure and housekeeping genes

A real-time thermocycler10 was used and quantification with external standards was performed with the fluorescent format for SYBR Green I dye as previously described (Waguespack et al. 2004a,b). Previously designed and created primers for IL-1β, IL-6, IL-8, COX-2, E-selectin and 3 housekeeping genes (β-actin, β2-microglobulin and glyceraldehyde-3-phosphate dehydrogenase) were used for RT-qPCR (Belknap et al. 2007; Loftus et al. 2007). All PCR reactions were performed in glass capillaries7 in 20 µl volumes (5 µl sample cDNA and 15 µl 1.33 X PCR master mixture). The master mixture was created and the PCR reaction was performed as per previously described for similar genes of interest (Belknap et al. 2007). Standard curves were performed for each gene product and water included as a negative control. A serial dilution of cDNA from each gene of interest was performed as previously described (Belknap et al. 2007) to generate standard curves. These curves were used for quantification of both the target genes and housekeeping genes in each cDNA sample for the normalisation process. Standards and laminar cDNA were prepared in separate capillaries but always amplified during the same PCR run. The reactions were performed in duplicate for the individual laminar cDNA samples from each horse.

Real-time qPCR was performed for 3 housekeeping genes. The resulting data for these genes were assessed by use of the geNorm computer program9. The 2 housekeeping genes that were reported as acceptable and received the best score from geNorm, β2-microglobulin and β-actin, were used to make a normalisation factor. The amplification data obtained by the RT-qPCR for the different genes were divided by the normalisation factor of the housekeeping genes in the same sample in order to normalise the RT-qPCR data for each gene of interest.


Once RT-qPCR sample data were divided by the normalisation factor, a D'Agostino and Pearson omnibus test was performed (which confirmed a normal distribution of the data sets), followed by a 2-tailed unpaired t test on the data of each individual gene of interest (IL-1β, IL-6, IL-8, COX-2, E-selectin) from each horse for the 2 treatment groups. Values of P<0.05 were considered significant. Laminar and skin leucocyte counts were tested for normal distribution with a D'Agostino and Pearson omnibus test, and a 2-tailed unpaired t test was performed between the 2 groups (SAL and LD). Values of P<0.05 were considered significant.


Physical examinations

The temperatures of all horses in the study ranged from 36.9–40.3°C. The average high temperature was higher (P = 0.0053) for the SAL group (39.4°C ± 0.26) vs. the LD group (37.8°C +/− 0.12). Two horses within the LD group exhibited signs of colic vs. 5 horses in the SAL group. All horses in the SAL group had decreased gastrointestinal sounds vs. 4 horses in the LD group. One horse (LD group) exhibited signs of ataxia lasting approximately 30–45 min in duration. Only one horse in the SAL group was subjectively considered “bright, alert and responsive” vs. 4 horses in the LD group.

Lidocaine concentrations

The lidocaine concentrations (mean ± s.e.) for the LD group at time points 0 h, 3 h and 10 h were 0 ± 0 µg/ml, 0.820 ± 0.117 µg/ml and 0.951 ± 0.192 µg/ml, respectively. Concentrations for Horse 6 (LD6) were well below the reported therapeutic minimum; therefore, the data for this horse were excluded from this study. Concentrations of all other horses in the study were within acceptable therapeutic ranges and thus included in the study.

Peripheral leucocyte counts

Mean ± s.e. white blood cell (WBC) count for all horses included in the study at time point 0 h (before administration of BWE or systemic saline/lidocaine) was 6.16 ± 0.39 × 109 cells/l. The mean ± s.e. WBC count for all horses included in the study at the 3 h and 10 h time point was 2.98 ± 0.28 × 109 cells/l and 7.83 ± 0.65 × 109 cells/l, respectively. There was a decrease (P<0.0001) in WBC count between the 0 h and 3 h time points and an increase (P<0.0001) in WBC count between the 3 h and 10 h time points for all horses receiving BWE. When comparing the averages of the WBC counts between the LD and SAL groups at these time points, there were no differences (0 h: P = 0.95, 3 h: P = 0.59 and 10 h: P = 0.81).

Leucocyte emigration

CD13-positive cells were present in the laminar interstitium of all horses at the 10 h time point (Fig 1) in numbers comparable to the results previously reported (Black et al. 2006). The number of cells present in the laminae of BWE treated horses at the 10 h time point in that study were increased (P<0.05) when compared to the laminae of the control (non-BWE treated) group. There was no difference (P = 0.91) between the number of CD13-positive cells in the laminae of horses treated with saline vs. lidocaine at the 10 h time point (Fig 1). As previously reported using CD13 immunohistochemistry to label skin leucocytes (Black et al. 2006), a small number of calprotectin-positive leucocytes were identified surrounding superficial dermal vessels in the skin immediately prior to BWE administration (0 h) in both the saline and lidocaine treated groups. There were increased numbers of calprotectinpositive cells surrounding the superficial dermal vessels in the skin at the 10 h time point in both the saline and lidocaine treated groups (Fig 2) when compared to the 0 h samples. There was no difference (P = 0.28 and P = 0.81) between the number of calprotectin-positive leucocytes in the skin dermis of horses treated with saline vs. lidocaine at either the 3 h or 10 h time points, respectively (Fig 2).

Figure 1.

Cell counts of CD13-positive cells (leucocytes) in laminae of salinetreated (SAL) and lidocaine-treated (LD) horses 10 h after nasogastric administration of black walnut extract (BWE). The bar located within the middle of the each column represents the mean of the data set.

Figure 2.

Cell counts of calprotectin-positive cells (leucocytes) surrounding superficial dermal vessels in laminae of saline-treated (SAL) and lidocaine-treated (LD) horses 0, 3 and 10 h after nasogastric administration of black walnut extract (BWE). The bar located within the middle of the each column represents the mean of the data set.

Laminar gene expression

To ensure that a similar inflammatory response occurred in the laminae in the current BWE protocol as that reported previously in this model, we compared laminar IL-6 mRNA concentrations between the SAL group (saline CRI following BWE administration) of the current study, and non-BWE treated control animals from a previous study (archived samples from previous BWE protocol [animals administered water instead of BWE], n = 5). There was a 491-fold increase (P = 0.015) in mRNA concentration of IL-6 in the laminae of the SAL group (BWE-treated, administered i.v. saline instead of lidocaine) when compared to laminar samples from the archived non-BWE treated controls (administered water), indicating that the BWE administered in the present study effectively induced laminar inflammation. When comparing the mRNA concentrations of cytokines IL-1β, IL-6 and IL-8 in the laminae of SAL vs. LD groups, there was no difference (P = 0.3876, P = 0.3505 and P = 0.3596, respectively) (Figs 3a–c). There was no difference (P = 0.7668) in mRNA concentration of COX-2 in the laminae of SAL vs. LD groups (Fig 3d). There was an increase (P = 0.0345) in laminar mRNA concentration of the endothelial adhesion molecule E-selectin in the LD group as compared to the SAL group (Fig 3e).

Figure 3.

Laminar mRNA concentrations of IL-1β (a), IL-6 (b), IL-8 (c), COX-2 (d) and E-selectin (e) obtained via RT-qPCR from saline- (SAL) and lidocaine-treated (LD) horses 10 h after nasogastric administration of black walnut extract. The bar located within the middle of the each column represents the mean of the data set.


Lidocaine has been investigated as a possible anti-inflammatory medication in multiple human diseases, including studies assessing ischaemia/reperfusion injury and ulcerative colitis (Bjorck et al. 2002; Yokoyama and Onishi 2005). In several studies, there is agreement that lidocaine modifies the normal immune response of a variety of cell types (Hollmann et al. 2001; Lan et al. 2005). The collective results of numerous in vitro studies evaluating the effects of lidocaine or lidocaine-related local anaesthetics on leucocytes indicate an inhibitory effect of lidocaine on neutrophil expression of surface integrins CD11b and CD18 (important in adhesion to endothelial surface, Martinsson et al. 1997), adhesion of leucocytes to injured venules (Giddon and Lindhe 1972; MacGregor et al. 1980; Martinsson et al. 1997), tissue migration of macrophages (Dickstein et al. 1985), and LPS-stimulated secretion of IL-1α and leucotriene B4 from peripheral blood mononuclear cells (Sinclair et al. 1993).

Furthermore, in an in vitro study by Lan et al. (2005), lidocaine attenuated the expression of cytokines/chemokines IL-1β, IL-6 and IL-8, as well as the expression of ICAM-1 (intracellular adhesion molecule-1) by activated endothelial cells (HUVECs). These antiinflammatory effects of lidocaine are further supported by the inhibition of secretion of IL-8 and IL-1β by TNF-α-stimulated intestinal epithelial cells exposed to lidocaine (Lahav et al. 2002). In vivo effects of lidocaine were examined in rabbits in which lidocaine attenuated increases in serum concentration of IL-8 and IL-6 following i.v. administration of endotoxin (Taniguchi et al. 2000).

In horses, lidocaine CRI has been assessed as a treatment for gastrointestinal injury using an intestinal mucosal healing model of equine small intestinal ischaemia/reperfusion (Cook et al. 2008). In this study, the effects of lidocaine with and without concurrent treatment with flunixin meglumine were examined on the intestinal wall recovery following ischaemic injury to the gastrointestinal tract. The results of the study indicated that systemic lidocaine attenuates a flunixin meglumine-induced increase in LPS permeability of ischaemic-injured jejunum (Cook et al. 2008). In a more recent report by the same investigators, the attenuation does not appear to be a direct inhibition of inflammatory events, as lidocaine alone had no effect on ischaemia-induced neutrophil extravasation (Cook et al. 2009a). It is possible that lidocaine's effect is only to limit sodium influx into ischaemia-injured cells (Sheu and Lederer 1985); excess intracellular sodium due to Na/K ATPase pump dysfunction in ischaemic epithelial cells reportedly disrupts formation of tight junctions and, therefore, the reestablishment of barrier function in an injured epithelial lining (Rajasekaran and Rajasekaran 2003). A more rapid return of the mucosal barrier function due to lidocaine-mediated change in intracellular sodium would not only decrease mucosal permeability, but would also remove a stimulus for neutrophil extravasation as the injured tissue returned to a normal state.

In equine medicine, a constant rate i.v. infusion of lidocaine is commonly used as a therapy to enhance motility or prevent the development of ileus in cases recovering from gastrointestinal surgery (Brianceau et al. 2002; Malone et al. 2006). However, despite its use as a pro-motility agent, the mechanisms of action of systemic lidocaine have not yet been elucidated. The proposed roles of systemic lidocaine that may contribute to its use for motility are that of an analgesic, a direct prokinetic, or an antiinflammatory agent (Cook and Blikslager 2008). The effectiveness of systemic lidocaine as an analgesic for gastrointestinal pain was challenged and determined to have no effect on visceral pain (Robertson et al. 2005). When tested as a direct prokinetic, lidocaine had no effect on contraction of the pylorus or midjejunum when placed topically (Nieto et al. 2000), and did not have a different effect on the myoelectrical activity of post operative jejunum as compared to the administration of saline when administered systemically (Milligan et al. 2007). Therefore, it has been suggested that the reported promotility effects of systemic lidocaine are due to its potential as an anti-inflammatory agent (Cook and Blikslager 2008). As discussed above, the results of the current study do not support any efficacy of lidocaine as a systemic anti-inflammatory therapy in the horse, and question any direct anti-inflammatory effect in any tissue including the GI tract.

Anecdotally, systemic lidocaine has been used clinically for equine cases suffering from laminitis (Belknap 2005). This action theoretically has merit given the aforementioned results of lidocaine and inflammation and more recent research that has indicated that similar to organ failure in human sepsis, early inflammation may lead to a series of events that culminate in failure of the laminae (Belknap et al. 2007). In the current study, laminar IL-1β, IL-6 and IL-8 concentrations were assessed due to the facts that: 1) these cytokines/chemokines are commonly increased in human sepsis studies (Cohen 2002; Boontham et al. 2003; Bhatia and Moochhala 2004; Frantz et al. 2005); 2) the 3 cytokines/chemokines have been found to be consistently increased in laminae in the BWE model of laminitis at a similar time point (Belknap et al. 2007); and 3) the 3 cytokines have been reported to be significantly decreased in studies using i.v. lidocaine (Lahav et al. 2002; Lan et al. 2005). Although TNF-α is also usually increased in human sepsis-related inflammation studies, it was not assessed as it has never been increased in laminar tissues in laminitis models (Belknap et al. 2007; Loftus et al. 2007).

There were no differences found in laminar mRNA concentrations of IL-1β, IL-6 and IL-8 between the LD and SAL groups, indicating no effect of a lidocaine CRI on expression of these inflammatory mediators. However, it was necessary to confirm that the lack of difference was not due to a lack of response of all animals to the BWE that was administered. Therefore, the increased laminar IL-6 mRNA concentrations in the SAL group from the current study (administered BWE followed by a CRI of saline) compared to archived non-BWE treated control laminar samples documented that the horses in this study underwent a similar inflammatory response as previously reported in this model. Furthermore, all horses receiving BWE in this study (SAL and LD groups) became leucopenic, a response consistent with systemic inflammation in the BWE model. Therefore, the lack of difference in cytokine mRNA concentrations between SAL and LD groups in the current study are due to a lack of efficacy of lidocaine, not due to a lack of inflammatory response of the horses to the BWE administration. The lack of effect of lidocaine in decreasing laminar IL-1β and IL-6 mRNA concentrations is corroborated by a previous study in HUVECs in which lidocainemediated attenuation of IL-1β and IL-6 gene expression by activated endothelial cells was only achieved at doses exceeding the toxic in vivo threshold and thus deemed clinically unsafe (Lan et al. 2005). At doses within the therapeutic range, lidocaine had no effect on the expression of IL-1β or IL-6 (Lan et al. 2005).

An important event in human sepsis-related organ injury that is also present in the equine laminae is a prominent emigration of leucocytes through an activated endothelium into the tissue interstitium (Cohen 2002; Mizgerd 2002; Strassheim et al. 2002; Annane et al. 2005; Nourshargh and Marelli-Berg 2005; Black et al. 2006). In human sepsis, leucocytes entering the extravascular tissues produce multiple cytokines, reactive oxygen species and matrix metalloproteases (MMPs), which ultimately can result in organ failure (Maier 2000; Aldridge 2002; Annane et al. 2005). Therefore, extravasated leucocytes are likely to be a primary source of injurious inflammatory molecules reported to be present in affected laminae including matrix metalloproteases and proinflammatory cytokines (Johnson et al. 1998; Fontaine et al. 2001; Belknap et al. 2004; Kyaw-Tanner and Pollitt 2004; Waguespack et al. 2004b; Black et al. 2006). Increased mRNA concentrations of the endothelial adhesion molecule, E-selectin, are present in affected laminae in the BWE model of laminitis at the time point laminar tissue was assessed in the current study (Loftus et al. 2007). E-selectin mediates the initial attachment/rolling of neutrophils and other leucocytes to venules at sites of inflammation (Abbas et al. 2000).

In the current study, there was an increase (P<0.05) in the expression of laminar E-selectin in the LD group compared to the SAL group; this was the only significant difference observed between LD and SAL groups in any of the parameters assessed in this study. Importantly, this increase in laminar E-selectin mRNA concentration in the LD group implies that the systemic administration of lidocaine has an inflammatory/activating effect on the endothelium. This in vivo finding is corroborated by a recent report in which ex vivo exposure of equine neutrophils to lidocaine did not inhibit neutrophil migration or adhesion at a therapeutic concentration, and actually increased neutrophil adhesion and transendothelial migration at higher concentrations of lidocaine (Cook et al. 2009b). Several properties of lidocaine have been investigated as possible causes of the drug's observed inflammatory effect. Lidocaine is an acidic local anaesthetic (Catterall and Mackie 2005), which could possibly be an irritant to cells due to its pH. However, the acidity of lidocaine was determined to not be a factor in its role as an initiator of inflammation in the in vitro study (Cook et al. 2009b).

Other factors, such as the presence of the preservative methyl paraben in 2% lidocaine solutions have been determined to be unlikely sources of the inflammatory effects seen at that concentration (Cook et al. 2009b). Although the current study does not define the reason for a proinflammatory effect of systemic lidocaine, the lidocaine-induced increase in laminar E-selectin expression does indicate that the reported ex vivo effect of lidocaine on adhesion and transendothelial migration of neutrophils may be due to activation of the endothelium. Finally, the similar laminar leucocyte counts in the LD and SAL groups in the current study indicate no inhibitory effect of lidocaine on overall leucocyte emigration (Black et al. 2006).

Skin biopsies were also assessed in the current study as they have a similar dermal/epidermal relationship as the laminae, and have been reported previously to reflect a similar increase in leucocyte emigration as that observed in the laminae (Black et al. 2006). Importantly, the skin represents a tissue not within the digit and therefore allows us to assess the effect of lidocaine CRI on systemic activation. Skin leucocyte counts were increased significantly at 10 h when compared to the 0 h time point for both groups; this finding is also consistent with that previously reported (Black et al. 2006). However, there was no difference in skin leucocyte counts between the LD and SAL group. Therefore, when comparing the SAL group to the LD group, there was no significant difference in leucocyte counts (and therefore most likely in leucocyte emigration) in either skin or laminae at any similarly compared time point.

Lidocaine appeared to have an antipyretic effect (significantly decreased maximum rectal temperature in LD vs. SAL horses) in the current study. Due to the lack of differences observed in all of the inflammatory parameters evaluated in this study, we do not believe that the lower temperatures observed in the LD group were due to systemic anti-inflammatory effects of lidocaine. Throughout the literature, there are reports of the effects of systemic lidocaine causing the opposite effect, hyperthermia (Tatsukawa et al. 1992; Crandall et al. 2002). It has been speculated that lidocaine-induced hyperthermia can occur due to suppressed uptake of calcium ions from the sarcoplasmic reticulum (Fujioka et al. 1988), leading to increased intracellular calcium concentration and muscular contraction (Britt 1974). However, other studies failed to induce hyperthermia in any of the animals administered i.v. lidocaine (Wingard and Bobko 1979; Harison and Morrell 1980). In one study, perfusion of the caudal hypothalamus with a sodium-rich perfusate resulted in a rise in body temperature (the opposite effect occurred with calcium) (Feldberg and Myers 1963); lidocaine may therefore block an effect such as this by limiting influx of sodium into the cellular constituents of the hypothalamus. Considering these data, it may be speculated that the temperature differences observed were due to a combination of effects of black walnut extract and lidocaine on the central thermoregulatory centres of the hypothalamus.

Plasma lidocaine concentrations were above the published therapeutic and below the toxic dose for all horses included in the LD group at the time of laminar sampling (10 h time point) (Meyer et al. 2001; Brianceau et al. 2002; Cook et al. 2008). The average lidocaine concentration for the LD group at the 3 h time point was slightly below the therapeutic minimum (0.82 µg/ml vs. 0.9 µg/ml) (Meyer et al. 2001; Brianceau et al. 2002; Cook et al. 2008). This is the only time in which samples were taken and evaluated for leucocyte emigration when lidocaine levels were below therapeutic levels. The reason for the slightly lower lidocaine concentrations at the 3 h time point is not known. However, as all horses included in the study received the standard clinical dose (1.3 mg/kg bwt bolus of lidocaine, followed by 0.05 mg/kg bwt/min constant rate infusion) controlled by titrated infusion pumps and signs of lidocaine toxicity were observed in one horse in the present study, it is likely that these are the concentrations commonly achieved in the clinical setting and, therefore, the data for the 3 h time point are still clinically relevant. The one horse in which the lidocaine plasma concentrations were below the appropriate therapeutic dose at sampling of the laminae due to an administration error was eliminated from the study. The elimination of this horse decreased the sample size of the LD group to 5 horses, a number used in previous laminitis studies (Belknap et al. 2007). As is the case with many equine studies, the strength of the study would be increased with a larger sample size. However, due to the overwhelming lack of difference between the 2 groups, there was not enough justification to add more horses to the study.

In summary, this study indicates that systemic lidocaine at concentrations achievable in the clinical setting does not have anti-inflammatory properties in regards to systemic inflammation documented in the black walnut extract model of laminitis. In fact, the data suggest that systemic lidocaine may possibly have a deleterious effect on systemic inflammatory disease by activating the endothelium. In conclusion, the present data suggest that lidocaine should not be used clinically for treatment of laminar inflammatory events that may lead to laminar destruction and failure in laminitis. Furthermore, when evaluating the laminar and skin results of this study in light of several other reports (discussed above) questioning the anti-inflammatory effects of lidocaine, lidocaine may not be an effective agent for addressing local or systemic inflammation in any species and its use as such a treatment is open to question.

Manufacturers' addresses

1 Schering Plough, Union, New Jersey, USA.

2 Abcam, Cambridge, Massachusetts, USA.

3 Vector Laboratories, Burlingame, California, USA.

4 Aperio Technologies, Vista, California, USA.

5 National Institutes of Health, Bethesda, Maryland, USA.

6 Waters Corporation, Milford, Massachusetts, USA.

7 Phenomenex, Torrance, California, USA.

8 Jasco, Easton, Maryland, USA.

9 Stratagende, Inc., LaJolla, California, USA.

10 Roche Molecular Biochemical, Indianapolis, Indiana, USA.

11 Ambion Inc., Austin, Texas, USA.

Author contributions The initiation, conception and planning for this study were by J.M.W., J.F.P., J.A.E.H. and J.K.B. Its execution was by J.M.W., Y.J.L., J.P.L., R.R.F., J.A.E.H. and J.K.B., with statistics by J.M.W. and J.K.B. The paper was written by J.M.W., R.R.F., W.R.R. and J.K.B.