Endotoxemia is a severe complication of many diseases in horses, such as acute gastrointestinal tract disorders, pulmonary and reproductive tract infections, and, in neonates, sepsis.1,2 The high morbidity and mortality rates associated with endotoxemia are attributed to the combined effects of a microbial pathogen and an animal's immune response. Systemic inflammatory response syndrome refers to activation and dysregulation of inflammatory pathways with secondary tissue necrosis and microvascular thrombosis that culminates in multiple organ failure.3–6 Experimental induction of endotoxemia by administration of a low dose of bacterial LPS has been used to evaluate the pathophysiology of sepsis-induced systemic inflammatory response syndrome.7–15 Several groups of investigators8,13,16,17 determined the upregulation and time course of leukocyte activation and inflammatory cytokine expression in equids that received LPS. Results of clinical studies18,19 indicate that inflammatory cytokine expression patterns of leukocytes obtained from horses with naturally occurring gastrointestinal tract disease are related to disease severity and outcome. Nevertheless, treatment with anti-inflammatory drugs is not highly effective for the management of horses with endotoxemia.2,20
Because of increasing recognition of a relationship between inflammation and coagulation, studies8,10,21,22 have been conducted to determine the effects of LPS on the activation status of platelets and hemostatic proteins. Administration of LPS to humans results in increased generation of thrombin by activation of the coagulation cascade with both tissue factor–dependent and –independent mechanisms.23,24 In addition, LPS promotes platelet-leukocyte interactions, which result in enhanced neutrophil tissue migration, platelet activation, and release of procoagulant microparticles from leukocytes, platelets, and endothelial cells.9,21,25–27
The purpose of the study reported here was to evaluate the effects of pretreatment with clopidogrel, a platelet inhibitory drug,28 on the clinicopathologic responses of horses to LPS administration. We hypothesized that administration of LPS to horses would induce platelet activation detectable with aggregation, viscoelastic, and flow cytometric assays and that pretreatment with clopidogrel would decrease clinicopathologic signs of inflammation in such horses.
Materials and Methods
Animals—Twelve healthy adult mares (5 Thoroughbreds, 4 Warmbloods, 2 Quarter Horses, and 1 Oldenburg) from an academic teaching facilitya were included in the study. The median age of the horses was 11.5 years (range, 8 to 25 years), and the median weight was 555 kg (range, 527 to 650 kg). Horses were determined to be healthy on the basis of histories and results of physical examinations, CBCs, and blood biochemical analyses. Prior to LPS administration, the horses had received either a placebo or clopidogrel for 72 hours; blood samples were obtained to determine the effects of clopidogrel on platelet function as part of another study.29 All study procedures were approved by an institutional laboratory animal care and use committee. The in vivo experiments and laboratory assays were performed between April and December 2011.
Study design and treatments—Blood samples (35 mL) for performance of baseline platelet function analyses were collected from each horse after a 24-hour acclimation period (3 days before the start of LPS administration). Horses then received clopidogrelb or a placebo (determined with a randomization procedure). Clopidogrel or placebo suspension was administered through a nasogastric tube followed by 750 mL of water to ensure gastric delivery of the entire dose. After an initial loading dose of clopidogrel (4 mg/kg), maintenance doses (2 mg/kg, q 24 h) were administered for 72 hours. One author (AEW) performed the randomization procedure and preparation of treatment suspensions. The other authors, who were unaware of the group assignments of horses, administered the treatments and performed physical examinations, blood sample collections, and laboratory analyses.
LPS administration—After collection of blood samples 72 hours after the start of clopidogrel or placebo administration, all horses received LPS (Escherichia coli O55:B5 endotoxinc [30 ng/kg, IV, during 30 minutes]), all horses received LPS (Escherichia coli O55:B5 endotoxinc [30 ng/kg, IV, during 30 minutes]). The lyophilized powder was reconstituted to maintain sterility in PBS solution to a concentration of 10 μg/mL and stored frozen in aliquots in siliconized glass containers until the day of administration. One hour prior to administration, an aliquot of the LPS preparation was thawed at room temperature (20° to 23°C) and then vortexed at room temperature for 1 hour. After vortexing, LPS (30 ng/kg) was added to 500 mL of saline (0.9% NaCl) solution and administered IV at a rate of 1,000 mL/h. The horses were examined, values of physical examination variables recorded (rectal temperature, heart rate, and respiration rate), and clinical comfort scores (on a scale of 1 through 3) determined.14 Blood samples were collected at the completion of administration of LPS (0 hours) and 1, 2, and 3 hours after that time. Another blood sample was collected from each horse 24 hours after the completion of administration of LPS. The horses were examined throughout the study period to detect signs of bleeding, petechiae, bruising, or swelling at venipuncture sites.
Blood sample collection and processing—Blood samples were collected by means of jugular venipuncture into evacuated glass tubesd containing no additive, EDTA, or heparin anticoagulant with a 17-gauge needle attached to a plastic catheter with a stopper-piercing needle.e The needle was then clipped off, a plastic syringe containing 2.0 mL of 3.8% sodium citrate was attached to the catheter tubing, and 18.0 mL of blood was gently withdrawn. Blood samples were processed as described.29 Briefly, EDTA-anticoagulated whole blood samples were used for CBCs, heparin-anticoagulated whole blood samples were used for modified thrombelastographic analyses, and citrated whole blood samples were used for kaolin-activated thrombelastographic analyses and determination of closure time with a tabletop whole blood platelet function analyzer.f Serum was harvested from whole blood samples without anticoagulant after incubation in a glass tube at 37°C for 20 minutes and centrifugation at 1,000 × g. Plateletrich plasma was prepared by centrifugation of citrated whole blood samples at 450 × g for 5 minutes. Plateletand leukocyte-rich plasma was harvested from aliquots of citrated whole blood samples that were allowed to settle for 20 minutes. Platelet-poor plasma was isolated by centrifugation of citrated whole blood samples at 14,500 × g for 1 minute; the supernatant plateletpoor plasma was stored at −50°C for subsequent batch testing. The heparinized and citrated blood samples were kept at room temperature throughout processing, which was complete within 1 hour after blood sample collection.
Laboratory analyses—The CBCs were performed at the Cornell University Clinical Pathology Laboratory with an automated processorg at each blood sample collection time. Platelet function assays were performed as described.29
Light transmission aggregometry was performed in an aggregometerh with platelet-rich plasma (≥ 150,000 platelets/μL) with no adjustment to a standard platelet count. The maximal percentage aggregation and area under the aggregation curve were measured in response to ADPi (5 and 10μM) and collagenj (6 μg/mL) stimulation. Aggregation analysis was performed at all blood sample collection times.
Closure time was measured at each blood sample collection time with a platelet function analyzerf and the manufacturer's reaction cartridges containing collagen and ADP.k The reactions were performed in duplicate and the mean values used for statistical analyses.
Thrombelastography was performed with a viscoelastic monitorl with kaolin-activatedm citrated whole blood samples, and the variables R (reaction time), K (clotting time), α angle, and MA (maximum amplitude) determined by use of the instrument software were recorded for data analysis. In addition, thrombelastography was performed to determine MA for heparin-anticoagulated whole blood samples activated with a reptilase reagent containing factor XIIIa and the reptilase reagent combined with an ADP agonist.m In an assay modification, the reconstituted reptilase reagent was diluted 1:4 in distilled water just before addition to the reaction mixture.29 Thrombelastography was performed for blood samples collected 3 days and 60 minutes before the start of LPS administration and 2 and 24 hours after the end of administration.
Flow cytometry was performed at all blood sample collection times for samples of platelet- and leukocyte-rich plasma dual labeled for a constitutively expressed platelet membrane antigen (CD61)n and 1 of 3 activation-dependent markers to detect membrane-bound fibrinogen,o P-selectin expression,p or phosphatidylserine externalization.q The subset of flow cytometry events with positive results for CD61 that were below the first decade of the log forward scatter scale was classified as small platelets and platelet membrane–derived microparticles. Acquisitionr was set for 7,500 platelet events defined by their forward and side scatter properties. The data were analyzeds to determine the percentage of platelets with positive results for CD61 and an activation marker and the number of platelet membrane–derived microparticles for each sample.
Fibrinogen concentration was measured with a functional Clauss assay,30 and VWF antigen concentration was measured with a quantitative ELISA.31 A pooled equine plasma sample (prepared from samples obtained from 10 healthy horses not included in this study) was used as an assay standard. The fibrinogen content of the standard was determined by a gravimetric method, and the VWF antigen concentration of the standard was assigned a value of 100%. Fibrinogen concentration was measured at all blood sample collection times, and VWF antigen concentration was measured for samples collected at 3 days and 60 minutes before the start of endotoxin administration and 2 and 24 hours after the end of administration.
The concentration of TNF-α was measured in serum samples for all blood sample collection times by use of a commercially available kitt to configure an ELISA. The kit contained capture and sandwich antibodies specific for equine TNF-α and an equine recombinant TNF-α assay standard. The manufacturer's protocol for microtiter plate coating and sample dilution were modified as described32 to optimize assay performance. In particular, the serum samples were diluted in a commercially available, proprietary bufferu intended to improve assay sensitivity and minimize matrix effects. In preliminary experiments performed with serum samples obtained from 3 healthy horses (not included in the present study), sample dilutions of 1:64 and 1:128 yielded linear recovery of recombinant equine TNF-α added to the samples in the range of 7.0 to 1,000 pg/mL, with a background TNF-α concentration < 10 pg/mL (data not shown). The serum samples for horses in the present study were assayed in duplicate at a 1:64 dilution.
Statistical analysis—Data were imported into a commercially available statistical software programv for analysis. Repeated-measures ANOVA was used to compare temporal changes in values of the clinical and laboratory analysis outcome variables for the 2 groups, with horse considered a random effect. This analysis was performed with a mixed procedure in the statistical software,v and an autoregressive correlation structure was specified. Group, time point, and their interaction were included as factors in the ANOVA. The Tukey-Kramer method was used to adjust for multiple pairwise comparisons. Values were evaluated across all time points and between 60 minutes before the start of LPS administration and the time of maximum endotoxin response (defined as the leukocyte count nadir), which occurred in all horses 60 minutes after the end of LPS administration. For all analyses, values of P ≤ 0.05 were considered significant.
Results
Clinical signs—Prior to LPS administration, all horses were healthy and values of physical examination variables were unremarkable. None of the clopidogrel-treated horses developed petechiae or ecchymoses, had abnormal bleeding from venipuncture sites, or had hemorrhage associated with passage of the nasogastric tube for drug administration. Before LPS administration, values of clinical variables were similar between clopidogrel and placebo group horses. Following endotoxin administration, horses in both groups had similar temporal patterns of increased heart rates, respiratory rates, rectal temperatures, and clinical comfort scores. However, the increase in heart rate from before LPS administration to 60 minutes after administration was only significant (P = 0.02) for the placebo group (Figure 1; Table 1). Similarly, there was a significant (P = 0.05) increase (albeit marginal) in respiratory rate from before LPS administration to 60 minutes after administration for the placebo group but not the clopidogrel group.
Mean ± SD heart rate (A), respiratory rate (B), rectal temperature (C), and clinical comfort score (D) at various times before and after IV administration of LPS to 12 adult mares that received clopidogrel (white circles; n = 6) or a placebo (black circles; 6). The line above each x-axis indicates the time of LPS administration. Times indicated on the x-axis refer to days (d), minutes (m), or hours (h). Notice that the x-axis scale is not linear.
Citation: American Journal of Veterinary Research 75, 8; 10.2460/ajvr.75.8.760
Mean ± SD heart rate (A), respiratory rate (B), rectal temperature (C), and clinical comfort score (D) at various times before and after IV administration of LPS to 12 adult mares that received clopidogrel (white circles; n = 6) or a placebo (black circles; 6). The line above each x-axis indicates the time of LPS administration. Times indicated on the x-axis refer to days (d), minutes (m), or hours (h). Notice that the x-axis scale is not linear.
Citation: American Journal of Veterinary Research 75, 8; 10.2460/ajvr.75.8.760
Mean ± SD heart rate (A), respiratory rate (B), rectal temperature (C), and clinical comfort score (D) at various times before and after IV administration of LPS to 12 adult mares that received clopidogrel (white circles; n = 6) or a placebo (black circles; 6). The line above each x-axis indicates the time of LPS administration. Times indicated on the x-axis refer to days (d), minutes (m), or hours (h). Notice that the x-axis scale is not linear.
Citation: American Journal of Veterinary Research 75, 8; 10.2460/ajvr.75.8.760
Mean ± SD values of clinical and laboratory analysis variables for 6 horses that received clopidogrel and 6 horses that received a placebo before and at the time of leukocyte nadir after IV administration of LPS.
| Variable | Group | Before LPS administration | After LPS administration at the time of leukocyte nadir | Adjusted P value* |
|---|---|---|---|---|
| Heart rate (beats/min) | Clopidogrel | 37.3 ± 3.3 | 56.0 ± 15.4 | 0.07 |
| Placebo | 37.0 ± 5.3 | 61.3 ± 10.0 | 0.02 | |
| Respiratory rate (breaths/min) | Clopidogrel | 15.2 ± 3.0 | 18.0 ± 5.5 | 0.9 |
| Placebo | 17.7 ± 5.4 | 28.7 ± 12.2 | 0.05 | |
| Temperature (°C) | Clopidogrel | 37.28 ± 0.22 | 37.94 ± 0.22 | < 0.001 |
| Placebo | 37.50 ± 0.33 | 37.89 ± 0.33 | 0.006 | |
| Hct (%) | Clopidogrel | 39.9 ± 4.8 | 39.2 ± 3.8 | — |
| Placebo | 40.1 ± 3.2 | 40.1 ± 2.6 | — | |
| WBC count (× 103 WBCs/μL) | Clopidogrel | 8.2 ± 0.7 | 2.2 ± 0.5 | < 0.001 |
| Placebo | 8.0 ± 1.0 | 2.3 ± 0.6 | < 0.001 | |
| Neutrophil count (× 103 neutrophils/μL) | Clopidogrel | 5.3 ± 0.6 | 1.0 ± 0.5 | < 0.001 |
| Placebo | 4.5 ± 0.6 | 0.8 ± 0.2 | < 0.001 | |
| Lymphocyte count (× 103 lymphocytes/μL) | Clopidogrel | 2.2 ± 0.3 | 1.0 ± 0.2 | < 0.001 |
| Placebo | 2.8 ± 0.8 | 1.3 ± 0.6 | < 0.001 | |
| Monocyte count (× 103 monocytes/μL) | Clopidogrel | 0.3 ± 0.1 | 0.01 ± 0.01 | < 0.001 |
| Placebo | 0.3 ± 0.1 | 0.02 ± 0.01 | < 0.001 | |
| Platelet count (× 103 platelets/μL) | Clopidogrel | 162.7 ± 24.3 | 136.7 ± 22.3 | 0.006 |
| Placebo | 153.2 ± 21.5 | 129.2 ± 16.6 | 0.01 | |
| Blood glucose (mg/dL) | Clopidogrel | 104.7 ± 10.7 | 107.0 ± 6.3 | — |
| Placebo | 108.3 ± 20.6 | 102.8 ± 10.2 | — | |
| Fibrinogen (mg/dL) | Clopidogrel | 384.7 ± 97.4 | 383.8 ± 136.1 | — |
| Placebo | 411.5 ± 38.5 | 447.3 ± 41.3 | — | |
| TNF-α (pg/mL) | Clopidogrel | 9,522.8 ± 9,847.3 | 13,004.0 ± 9,708.7 | 0.06 |
| Placebo | 7,042.7 ± 8,814.3 | 9,918.2 ± 8,481.4 | 0.08 | |
| Closure time (s) | Clopidogrel | 89.0 ± 18.7 | 93.5 ± 16.0 | — |
| Placebo | 93.1 ± 15.6 | 91.7 ± 23.3 | — | |
| ADP-induced aggregation (%) | ||||
| 5μM | Clopidogrel | 39.5 ± 24.1 | 23.8 ± 16.6 | 0.05 |
| Placebo | 60.2 ± 9.0 | 47.4 ± 8.4 | 0.1 | |
| 10μM | Clopidogrel | 45.7 ± 22.4 | 37.5 ± 22.8 | 0.3 |
| Placebo | 63.2 ± 5.3 | 53.0 ± 7.4 | 0.2 | |
| CD62P (%) | Clopidogrel | 2.6 ± 2.2 | 0.7 ± 0.3 | 0.03 |
| Placebo | 1.1 ± 0.3 | 0.5 ± 0.2 | 0.7 | |
| Annexin V (%) | Clopidogrel | 3.0 ± 5.5 | 0.6 ± 0.3 | — |
| Placebo | 0.9 ± 0.3 | 0.7 ± 0.4 | — | |
| VWF antigen (%) | Clopidogrel | 143.2 ± 42.9 | 234.5 ± 55.1 | 0.01 |
| Placebo | 154.2 ± 20.1 | 168.8 ± 37.3 | 0.9 | |
| R (min) | Clopidogrel | 19.9 ± 2.7 | 15.3 ± 2.0 | 0.005 |
| Placebo | 18.0 ± 2.4 | 14.3 ± 3.9 | 0.02 | |
| K (min) | Clopidogrel | 6.5 ± 2.5 | 4.0 ± 0.6 | 0.05 |
| Placebo | 6.2 ± 0.8 | 4.1 ± 2.0 | 0.1 | |
| α angle (°) | Clopidogrel | 25.5 ± 8.7 | 26.5 ± 11.5 | — |
| Placebo | 20.9 ± 2.5 | 25.7 ± 7.4 | — | |
| MA (mm) | ||||
| Kaolin activated | Clopidogrel | 60.8 ± 11.0 | 50.4 ± 5.4 | 0.05 |
| Placebo | 58.3 ± 6.2 | 51.9 ± 6.1 | 0.2 | |
| Reptilase activated | Clopidogrel | 28.6 ± 9.5 | 22.5 ± 5.9 | 0.4 |
| Placebo | 39.5 ± 7.9 | 18.8 ± 11.0 | 0.002 | |
| Reptilase with ADP activated | Clopidogrel | 34.0 ± 8.3 | 28.4 ± 8.4 | 0.5 |
| Placebo | 42.5 ± 7.5 | 26.8 ± 9.2 | 0.008 | |
Adjusted P values determined with post hoc comparisons are reported for data with significant (P ≤ 0.05) repeated-measures ANOVA results.
— = Not reported. K = Clotting time. R = Reaction time.
CBCs—Prior to LPS administration, there were no significant differences in blood cell counts between groups. By 60 minutes after LPS administration (Table 1), both groups had a marked and significant (P < 0.001) decrease in leukocyte count, with a similar degree and duration of neutropenia and monocytopenia. Clopidogrel (P = 0.006) and placebo (P = 0.01) groups also had significant decreases in platelet counts 60 minutes after LPS administration, compared with values before administration. However, no significant differences were found between groups for any blood cell count after LPS administration (Figure 2).
Mean ± SD total leukocyte count (A), neutrophil count (B), monocyte count (C), and platelet count (D) at various times before and after IV administration of LPS to 12 adult mares that received clopidogrel (white circles; n = 6) or a placebo (black circles; 6). The line above each x-axis indicates the time of LPS administration. See Figure 1 for remainder of key.
Citation: American Journal of Veterinary Research 75, 8; 10.2460/ajvr.75.8.760
Mean ± SD total leukocyte count (A), neutrophil count (B), monocyte count (C), and platelet count (D) at various times before and after IV administration of LPS to 12 adult mares that received clopidogrel (white circles; n = 6) or a placebo (black circles; 6). The line above each x-axis indicates the time of LPS administration. See Figure 1 for remainder of key.
Citation: American Journal of Veterinary Research 75, 8; 10.2460/ajvr.75.8.760
Mean ± SD total leukocyte count (A), neutrophil count (B), monocyte count (C), and platelet count (D) at various times before and after IV administration of LPS to 12 adult mares that received clopidogrel (white circles; n = 6) or a placebo (black circles; 6). The line above each x-axis indicates the time of LPS administration. See Figure 1 for remainder of key.
Citation: American Journal of Veterinary Research 75, 8; 10.2460/ajvr.75.8.760
Platelet aggregation—For horses that received clopidogrel, a significant decrease in the platelet aggregation response was detected in blood samples stimulated with 5 and 10 μM ADP.29 However, that inhibitory effect varied among horses. Two horses had a small or no decrease in platelet aggregation, whereas 4 horses had > 50% inhibition of platelet aggregation, compared with the baseline value. Although a further decrease in platelet aggregation was detected from 1 to 3 hours after LPS administration for horses that received clopidogrel, the difference between these time points was not significant, even when data for the 2 horses without a substantial decrease in platelet aggregation were removed from the analysis (Figure 3; Table 1). Administration of LPS did not cause significant changes in platelet aggregation in response to addition of ADP or collagen to blood samples for horses in the placebo group.
Percentage of maximal platelet aggregation in response to ADP agonist treatment (10μM) of blood samples collected from adult mares at various times before and after IV administration of LPS. A—Mean ± SD values for all horses in the study that received clopidogrel (white circles; n = 6) or a placebo (black circles; 6). B—Mean ± SD values for horses that received the placebo (black circles; n = 6) and horses treated with clopidogrel that responded to that treatment (as determined by a substantial decrease in platelet aggregation; white circles; 4). The line above each x-axis indicates the time of LPS administration. See Figure 1 for remainder of key.
Citation: American Journal of Veterinary Research 75, 8; 10.2460/ajvr.75.8.760
Percentage of maximal platelet aggregation in response to ADP agonist treatment (10μM) of blood samples collected from adult mares at various times before and after IV administration of LPS. A—Mean ± SD values for all horses in the study that received clopidogrel (white circles; n = 6) or a placebo (black circles; 6). B—Mean ± SD values for horses that received the placebo (black circles; n = 6) and horses treated with clopidogrel that responded to that treatment (as determined by a substantial decrease in platelet aggregation; white circles; 4). The line above each x-axis indicates the time of LPS administration. See Figure 1 for remainder of key.
Citation: American Journal of Veterinary Research 75, 8; 10.2460/ajvr.75.8.760
Percentage of maximal platelet aggregation in response to ADP agonist treatment (10μM) of blood samples collected from adult mares at various times before and after IV administration of LPS. A—Mean ± SD values for all horses in the study that received clopidogrel (white circles; n = 6) or a placebo (black circles; 6). B—Mean ± SD values for horses that received the placebo (black circles; n = 6) and horses treated with clopidogrel that responded to that treatment (as determined by a substantial decrease in platelet aggregation; white circles; 4). The line above each x-axis indicates the time of LPS administration. See Figure 1 for remainder of key.
Citation: American Journal of Veterinary Research 75, 8; 10.2460/ajvr.75.8.760
Closure time—No significant differences were found between clopidogrel and placebo groups regarding closure time in blood samples obtained before LPS administration. Neither group had a significant change in closure time after LPS administration (Table 1).
Thrombelastography—No significant differences were found between clopidogrel and placebo groups before LPS administration for any of the kaolin-activated thrombelastography variables (R, K, α angle, and MA) or for MA in reptilase-activated and reptilase with ADP-activated heparinized blood samples. After LPS administration (Table 1), both groups had a faster rate of clot formation (R and K) and lower clot strength (MA for kaolin-activated citrated, reptilase-activated, and reptilase with ADP-activated heparinized blood) compared with values before administration. However, the decrease in clot strength (MA) between blood samples obtained before LPS administration and those obtained 120 minutes after administration was only significant for the placebo group for reptilase-activated blood samples (P = 0.002) and reptilase with ADP-activated blood samples (P = 0.008).
Platelet flow cytometry—No significant differences were detected for the placebo group after LPS administration regarding the percentage of platelets with membrane-bound fibrinogen, P-selectin expression, or phosphatidylserine externalization or in the number of platelet membrane–derived microparticles compared with values before administration. Of these platelet activation markers, only P-selectin expression was significantly (P = 0.01) different between groups (Figure 4). For the clopidogrel group, there was a significant (P = 0.03) decrease in the percentage of platelets with positive results for P-selectin 60 minutes after LPS administration, compared with baseline values (Table 1). This decrease in value was no longer detected by 24 hours after LPS administration.
Percentage of platelets with expression of P-selectin (on the basis of cytometric analyses of 7,500 platelet events; A), plasma fibrinogen concentration (B), plasma VWF antigen concentration (C), and serum TNF-α concentration (D) at various times before and after IV administration of LPS to 12 adult mares that received clopidogrel (white circles; n = 6) or a placebo (black circles; 6). The line above each x-axis indicates the time of LPS administration. See Figure 1 for remainder of key.
Citation: American Journal of Veterinary Research 75, 8; 10.2460/ajvr.75.8.760
Percentage of platelets with expression of P-selectin (on the basis of cytometric analyses of 7,500 platelet events; A), plasma fibrinogen concentration (B), plasma VWF antigen concentration (C), and serum TNF-α concentration (D) at various times before and after IV administration of LPS to 12 adult mares that received clopidogrel (white circles; n = 6) or a placebo (black circles; 6). The line above each x-axis indicates the time of LPS administration. See Figure 1 for remainder of key.
Citation: American Journal of Veterinary Research 75, 8; 10.2460/ajvr.75.8.760
Percentage of platelets with expression of P-selectin (on the basis of cytometric analyses of 7,500 platelet events; A), plasma fibrinogen concentration (B), plasma VWF antigen concentration (C), and serum TNF-α concentration (D) at various times before and after IV administration of LPS to 12 adult mares that received clopidogrel (white circles; n = 6) or a placebo (black circles; 6). The line above each x-axis indicates the time of LPS administration. See Figure 1 for remainder of key.
Citation: American Journal of Veterinary Research 75, 8; 10.2460/ajvr.75.8.760
Fibrinogen and VWF antigen concentrations—No significant differences in plasma fibrinogen concentration were detected between groups after LPS administration. Although plasma fibrinogen concentrations 24 hours after LPS administration were higher for the placebo and clopidogrel groups (474 and 443 mg/dL, respectively) than they were for those groups before administration (approx 412 and 385 mg/dL, respectively), these differences were not significant for either group (Table 1; Figure 4). Similarly, plasma VWF antigen concentrations for the placebo and clopidogrel groups 24 hours after LPS administration (190% and 165%, respectively) were higher than the values before administration (146% and 137%, respectively). The change in this value over time varied by group owing to a significant (P = 0.01) increase at 120 minutes after LPS administration, compared with values before administration, for the clopidogrel group but not the placebo group.
TNF-α—A wide range of baseline serum TNF-α concentrations were detected among horses (range, 7 to 23,138 pg/mL; mean, 8,583 pg/mL; Figure 4). Although a wide range of serum TNF-α concentrations measured with ELISAs have been reported32–34 for horses, 6 of 12 horses in the present study (3 in the placebo group and 3 in the clopidogrel group) had concentrations higher than the highest reported reference range cutoff value (5,000 pg/mL). Both groups of horses in this study had a qualitatively similar serum TNF-α response to LPS administration, with an early peak in the mean serum TNF-α concentration 60 minutes after LPS administration and a return to concentrations similar to baseline values by 24 hours after administration. However, neither group had a significant increase in serum TNF-α concentration after LPS administration (Table 1).
Discussion
Results of this study indicated no significant changes in measures of platelet aggregation, adhesion, degranulation, or procoagulant activity for healthy horses that received LPS, despite development of profound leukopenia and clinical signs associated with cytokine release. However, treatment with the antiplatelet drug clopidogrel before administration of LPS resulted in apparent attenuation of the clinical responses to endotoxin and transient changes in platelet P-selectin expression and plasma VWF antigen concentration. In contrast to the placebo group, the clopidogrel group had no significant increases in mean heart rate after LPS administration. Of the other variables evaluated, the clopidogrel group had a significant decrease in platelet P-selectin expression within 1 hour after LPS administration and an increase in plasma VWF concentration by 2 hours after administration. Although the mean platelet aggregation values after clopidogrel treatment for horses that received the drug were significantly lower than values before treatment, 2 of the 6 treated horses seemed to be refractory to the antiaggregation action of the drug. Combined, these findings indicated that administration of LPS to horses in this study did not induce a measurable increase in the hemostatic properties of circulating platelets and suggested that potential clinical benefits of clopidogrel in endotoxemic horses would be independent of its antiaggregatory action. Differences in clinical responses detected among horses may have been attributable to clopidogrel's inhibition of signaling through P2Y12 receptors on inflammatory cells or nonhematopoietic cell types.35,36
Administration of LPS is a well-established method of experimental induction of endotoxemia in horses.8,10,13,15,17,37 Lipopolysaccharide causes an innate immune response that is primarily mediated by neutrophils, monocytes, and tissue macrophages.9,13–16,24,26 Leukocyte activation is initiated when LPS, bound to accessory proteins such as CD14 and LPS-binding protein, interacts with the transmembrane receptor toll-like receptor-4, resulting in downstream signaling events, translocation of transcription factors to the nucleus, and subsequent increases in protein synthesis. Plasma concentrations of TNF-α are among the first inflammatory cytokine concentrations to increase as a result of this signaling pathway. Peak TNF-α concentrations are typically detected within 60 minutes after LPS administration, at the same time that systemic signs of an inflammatory response become apparent.13–15,17,38
The horses in this study had a characteristic time course of clinical signs and cytopenia in response to LPS administration. In another study8 of healthy horses, the dose of LPS that was administered to horses in the present study induced a rapid increase in platelet p38 MAPK phosphorylation with a return to baseline values by 60 minutes after administration and an increase in circulating concentrations of thromboxane and serotonin. In the present study, we included additional measures of platelet activation such as measures of adhesion, aggregation, α-granule release, procoagulant activity, and viscoelastic measures of clot strength. Results of these analyses indicated no evidence of clinically relevant increases in platelet reactivity following LPS administration. Similar to results of another study10 in which a different instrument was used to measure viscoelastic properties of blood samples, we found that the values of MA reflecting overall clot stability or firmness were decreased after administration of LPS. Although platelets express toll-like receptor-4 receptors, their response to LPS treatment is complex and highly influenced by reaction conditions.8,23,24,26,27,39–42 Ex vivo incubation of platelets with LPS at concentrations on the order of micrograms per milliliter induces degranulation and enhances aggregation and phosphatidylserine expression; however, such concentrations are several log-fold higher than the concentrations attained after administration of LPS to horses or those attained secondary to clinical diseases, which are on the order of picogram per milliliter.8,25,26,42
The presence and interaction of leukocytes and plasma proteins with platelets determine their response to LPS. An increase in platelet aggregation and degranulation requires leukocyte release of cytokines, platelet-activating factors, and soluble plasma CD14 that augment the effects of LPS alone.8,13,24,39,41,42 Together, these data suggest that the circulating LPS concentrations attained in horses with clinical disease and those in the present study may have no direct platelet agonist activity that stimulates aggregation; rather, LPS may initiate platelet signaling pathways and transcript processing that integrates platelet, leukocyte, and vascular endothelial cell activation into a coordinated immune response to an inflammatory stimulus.
We found that pretreatment with clopidogrel seemed to decrease some of the adverse systemic effects of LPS administration to horses. Clopidogrel, which is in the thienopyridine class of platelet inhibitors, requires hepatic conversion to an active metabolite that irreversibly binds to the platelet P2Y12 receptor.28,43 Blockade of this receptor impairs platelet response to ADP stimulation, and clopidogrel is commonly used alone or in combination with aspirin for its antiaggregatory action in humans with atherosclerosis and acute coronary syndromes. The efficacy of clopidogrel in such patients is related to prodrug metabolism by the hepatic cytochrome P450 system. Interindividual variation in generation of the active metabolite is the primary cause of poor responses to treatment in humans.44–47 Two of the 6 horses treated with clopidogrel in the present study had no significant decrease in ADP-induced platelet aggregation, which indicated potential problems with drug absorption or hepatic metabolism. Another study48 of 6 horses treated with clopidogrel (2 mg/kg/d for 5 days) included measurement of plasma concentrations of an inactive clopidogrel metabolite (SR 26334). Peak plasma concentration of the metabolite was detected 30 minutes after the first administration of the drug in all horses, indicating that the drug had been absorbed. However, detectable concentrations of this metabolite were found only in 1 horse after 24 hours, whereas significant decreases in ADP-induced platelet aggregation were sustained from days 3 through 5 after the first administration of the drug. Additional pharmacokinetic and pharmacodynamic studies would be required to determine the causes of such variations in responses of horses to clopidogrel administration.
The limitations of the present study included a small sample size and high interindividual variation in the platelet inhibitory action of clopidogrel that might have further compromised our ability to detect significant effects of the drug and other aspects of the LPS response. For example, the significant decrease in P-selectin expression after LPS administration for the clopidogrel group was attributed to high expression before LPS administration for 2 of the 6 horses. Data for additional horses with a wide range of baseline P-selectin expression values would be needed to determine whether LPS induces a preferential loss of platelets with P-selectin expression from the vascular space. The fact that both groups of horses in this study had a significant decrease in platelet counts from baseline values at the time of leukocyte nadir is consistent with platelet-mediated leukocyte tissue migration. Therefore, the platelet fraction available for sampling may have represented a refractory population. Detection of shed receptors, such as soluble CD40 ligand or granule contents such as soluble P-selectin, rather than platelet membrane–associated activation markers, may have improved our ability to characterize the platelet activation response.49 Low platelet counts may also have contributed to the decrease in clot strength after LPS administration, given that platelet numbers can influence MA, independent of platelet function.50
The increase in the circulating VWF antigen concentration for the clopidogrel group reflected an increase for all 6 treated horses. Although the increase was significant only for the clopidogrel group, an increase compared with the circulating concentration before LPS administration was detected for both groups of horses 2 hours after LPS administration. This early time and transient increase suggested a release of VWF from intracellular stores,51 rather than upregulation of synthesis during an acute phase response; however, additional studies would be needed to further characterize the effects of LPS or clopidogrel on plasma VWF concentrations.
Other limitations of the present study were the few blood sample collection times, compared with the number in other studies8,10,13,15 in which LPS was administered, and the focus on measurement of variables of platelet activation related to formation of a hemostatic plug. We selected the 60-minute postadministration time as representative of the maximal LPS effect and the time 24 hours after administration as representative of the delayed response to LPS. More significant differences between groups may have been detected if additional early and late blood sample collection times had been included. We did not directly determine leukocyte or endothelial cell activation. Despite our use of an assay buffer reported to eliminate matrix interference,32 the high background absorbance values in the TNF-α assays for 6 of the 12 horses in the study precluded our use of the circulating concentration of this cytokine to accurately determine the severity of the inflammatory response for horses in either group. The use of a TNF-α bioassay,8 instead of an ELISA, could have avoided matrix effects on assay specificity.
Results of this study indicated no evidence of platelet hyperreactivity in healthy horses following LPS administration, on the basis of assessments of platelet adhesion, aggregation, and procoagulant activity. Modulation of platelet interactions with inflammatory cells and potential inhibition of P2Y12 receptors on inflammatory cells may be more important treatment targets than modulation of the activation pathways promoting hemostatic plug formation. For example, inhibitors of platelet p38 MAPK do not impair platelet aggregation, yet signaling through p38 MAPK is implicated in the ability of platelets to activate leukocytes.8,23,39,40 Although inhibition of the proinflammatory actions of platelets might be desirable, platelets also support vascular integrity52 and results of a recent study42 indicate antibacterial effects of LPS-stimulated equine platelets. Clinical trials of various antiplatelet agents are therefore needed to determine whether their use improves outcomes for patients with sepsis. Results of the present study and those of other studies29,48,53 indicate that clopidogrel has good safety in horses. Clopidogrel's recent change to an off-patent status reduces drug costs so that its use is feasible in veterinary clinical practice. However, measures to identify patients with high platelet reactivity after clopidogrel treatment may be needed to better determine its efficacy in clinical trials and clinical practice.
ABBREVIATIONS
| LPS | Lipopolysaccharide |
| p38 MAPK | p38 Mitogen activated protein kinases |
| TNF | Tumor necrosis factor |
| VWF | von Willebrand factor |
Cornell University Equine Park, Ithaca, NY.
Plavix, clopidogrel 75 mg USP, Bristol-Meyers Squibb, Princeton, NJ.
LPS, Sigma-Aldrich, St Louis, Mo.
BD Vacutainer K2 EDTA and Vacutainer Heparin, BD, Franklin Lakes, NJ.
Blood Collection Set, Hospira Inc, Lake Forest, Ill.
PFA-100 System, Siemens Diagnostics, Tarrytown, NY.
Advia 2120, Siemens Diagnostics, Tarrytown, NY.
Model 500 CA, Lumi-Aggregometer, Chronolog Corp, Havertown, Pa.
ADP, Chronolog Corp, Havertown, Pa.
Collagen (type 1, equine tendon), Chronolog Corp, Havertown, Pa.
Col/ADP test cartridges, Siemens Diagnostics, Tarrytown, NY.
TEG 5000 Thromboelastograph Haemostasis Analyzer System, Haemonetics, Braintree, Mass.
PlateletMapping Assay, ADP, Haemonetics, Braintree, Mass.
CD 61PE (clone SZ21), Beckman Coulter Inc, Brea, Calif.
Fibrinogen-fluorescein isothiocyanate, Dako, Carpintiera, Calif.
CD62P-DY488 (clone Psel.KO.2.7), Novus Biologicals, Littleton, Colo.
Annexin V-fluorescein isothiocyanate, R&D Systems Inc, Minneapolis, Minn.
BD FACSCalibur, Becton Dickinson, San Jose, Calif.
FloJo, version 9.5.3, Treestar Inc, Ashland, Ore.
DuoSet ELISA Development kit for Equine TNF-α, R&D Systems Inc, Minneapolis, Minn.
Bio-Plex Pro Isotyping Diluent, Bio-Rad, Mississagua, ON, Canada.
SAS, version 9.2, SAS Institute Inc, Cary, NC.
References
1. Sykes BW, Furr MO. Equine endotoxaemia—a state-of-the-art review of therapy. Aust Vet J 2005; 83: 45–50.
2. Werners AH, Bull S, Fink-Gremmels J. Endotoxaemia: a review with implications for the horse. Equine Vet J 2005; 37: 371–383.
3. Angus DC, van der Poll T. Severe sepsis and septic shock. N Engl J Med 2013; 369: 840–851.
4. Huisse MG, Pease S & Hurtado-Nedelec M, et al. Leukocyte activation: the link between inflammation and coagulation during heatstroke. A study of patients during the 2003 heat wave in Paris. Crit Care Med 2008; 36: 2288–2295.
5. Joop K, Berckmans RJ & Nieuwland R, et al. Microparticles from patients with multiple organ dysfunction syndrome and sepsis support coagulation through multiple mechanisms. Thromb Haemost 2001; 85: 810–820.
6. Weber GF, Swirski FK. Immunopathogenesis of abdominal sepsis. Langenbecks Arch Surg 2014; 399: 1–9.
7. Berg RM, Taudorf S & Bailey DM, et al. Effects of lipopolysaccharide infusion on arterial levels and transcerebral exchange kinetics of glutamate and glycine in healthy humans. APMIS 2012; 120: 761–766.
8. Brooks AC, Menzies-Gow NJ & Wheeler-Jones C, et al. Endotoxin-induced activation of equine platelets: evidence for direct activation of p38 MAPK pathways and vasoactive mediator production. Inflamm Res 2007; 56: 154–161.
9. Guha M, O'Connell MA & Pawlinski R, et al. Lipopolysaccharide activation of the MEK-ERK1/2 pathway in human monocytic cells mediates tissue factor and tumor necrosis factor alpha expression by inducing Elk-1 phosphorylation and Egr-1 expression. Blood 2001; 98: 1429–1439.
10. McGovern KF, Lascola KM & Smith SA, et al. The effects of hyperglycemia and endotoxemia on coagulation parameters in healthy adult horses. J Vet Intern Med 2013; 27: 347–353.
11. Michaeli B, Martinez A & Revelly JP, et al. Effects of endotoxin on lactate metabolism in humans. Crit Care 2012; 16: R139.
12. Moore JN, Morris DD. Endotoxemia and septicemia in horses: experimental and clinical correlates. J Am Vet Med Assoc 1992; 200: 1903–1914.
13. Nieto JE, MacDonald MH & Braim AE, et al. Effect of lipopolysaccharide infusion on gene expression of inflammatory cytokines in normal horses in vivo. Equine Vet J 2009; 41: 717–719.
14. Poulin Braim AE, MacDonald MH & Bruss ML, et al. Effects of intravenous administration of pirfenidone on horses with experimentally induced endotoxemia. Am J Vet Res 2009; 70: 1031–1042.
15. Tadros EM, Frank N. Effects of continuous or intermittent lipopolysaccharide administration for 48 hours on the systemic inflammatory response in horses. Am J Vet Res 2012; 73: 1394–1402.
16. Morris DD. Endotoxemia in horses. A review of cellular and humoral mediators involved in its pathogenesis. J Vet Intern Med 1991; 5: 167–181.
17. Morris DD, Crowe N, Moore JN. Correlation of clinical and laboratory data with serum tumor necrosis factor activity in horses with experimentally induced endotoxemia. Am J Vet Res 1990; 51: 1935–1940.
18. Barton MH, Collatos C. Tumor necrosis factor and interleukin-6 activity and endotoxin concentration in peritoneal fluid and blood of horses with acute abdominal disease. J Vet Intern Med 1999; 13: 457–464.
19. Lopes MA, Salter CE & Vandenplas ML, et al. Expression of inflammation-associated genes in circulating leukocytes collected from horses with gastrointestinal tract disease. Am J Vet Res 2010; 71: 915–924.
20. Moore JN, Barton MH. Treatment of endotoxemia. Vet Clin North Am Equine Pract 2003; 19: 681–695.
21. Piccin A, Murphy WG, Smith OP. Circulating microparticles: pathophysiology and clinical implications. Blood Rev 2007; 21: 157–171.
22. Schouten M, Wiersinga WJ & Levi M, et al. Inflammation, endothelium, and coagulation in sepsis. J Leukoc Biol 2008; 83: 536–545.
23. Brown GT, McIntyre TM. Lipopolysaccharide signaling without a nucleus: kinase cascades stimulate platelet shedding of proinflammatory IL-1beta-rich microparticles. J Immunol 2011; 186: 5489–5496.
24. Soop A, Hallstrom L & Frostell C, et al. Effect of lipopolysaccharide administration on the number, phenotype and content of nuclear molecules in blood microparticles of normal human subjects. Scand J Immunol 2013; 78: 205–213.
25. Bailey SR, Adair HS & Reinemeyer CR, et al. Plasma concentrations of endotoxin and platelet activation in the developmental stage of oligofructose-induced laminitis. Vet Immunol Immunopathol 2009; 129: 167–173.
26. Kornerup KN, Salmon GP & Pitchford SC, et al. Circulating platelet-neutrophil complexes are important for subsequent neutrophil activation and migration. J Appl Physiol (1985) 2010; 109: 758–767.
27. Ouellette AL, Evans RJ, Heath MF. Platelets enhance endotoxininduced monocyte tissue factor (TF) activity in the horse. Res Vet Sci 2004; 76: 31–35.
28. Savi P, Herbert JM. Clopidogrel and ticlopidine: P2Y12 adenosine diphosphate-receptor antagonists for the prevention of atherothrombosis. Semin Thromb Hemost 2005; 31: 174–183.
29. Brooks MB, Divers TJ & Watts AE, et al. Effects of clopidogrel on the platelet activation response in horses. Am J Vet Res 2013; 74: 1212–1222.
30. Gaffney PJ, Wong MY. Collaborative study of a proposed international standard for plasma fibrinogen measurement. Thromb Haemost 1992; 68: 428–432.
31. Benson RE, Catalfamo JL, Dodds WJ. A multispecies enzyme-linked immunosorbent assay for von Willebrand's factor. J Lab Clin Med 1992; 119: 420–427.
32. Lavoie-Lamoureux A, Maghni K, Lavoie JP. Optimization of a procedure to accurately detect equine TNFalpha in serum samples. Vet Immunol Immunopathol 2010; 138: 118–123.
33. Suagee JK, Burk AO & Quinn RW, et al. Effects of diet and weight gain on circulating tumour necrosis factor-alpha concentrations in Thoroughbred geldings. J Anim Physiol Anim Nutr (Berl) 2011; 95: 161–170.
34. Vick MM, Adams AA & Murphy BA, et al. Relationships among inflammatory cytokines, obesity, and insulin sensitivity in the horse. J Anim Sci 2007; 85: 1144–1155.
35. Gachet C. P2Y(12) receptors in platelets and other hematopoietic and non-hematopoietic cells. Purinergic Signal 2012; 8: 609–619.
36. Jacob F, Perez Novo C & Bachert C, et al. Purinergic signaling in inflammatory cells: P2 receptor expression, functional effects, and modulation of inflammatory responses. Purinergic Signal 2013; 9: 285–306.
37. Semrad SD, Moore JN. Effects of a specific thromboxane synthetase inhibitor in equine endotoxaemia. Res Vet Sci 1987; 43: 137–142.
38. Moore JN, Norton N & Barton MH, et al. Rapid infusion of a phospholipid emulsion attenuates the effects of endotoxaemia in horses. Equine Vet J 2007; 39: 243–248.
39. Brooks AC, Menzies-Gow NJ & Wheeler-Jones C, et al. Regulation of platelet activating factor-induced equine platelet activation by intracellular kinases. J Vet Pharmacol Ther 2009; 32: 189–196.
40. Shashkin PN, Brown GT & Ghosh A, et al. Lipopolysaccharide is a direct agonist for platelet RNA splicing. J Immunol 2008; 181: 3495–3502.
41. Zhang G, Han J & Welch EJ, et al. Lipopolysaccharide stimulates platelet secretion and potentiates platelet aggregation via TLR4/MyD88 and the cGMP-dependent protein kinase pathway. J Immunol 2009; 182: 7997–8004.
42. Aktan I, Dunkel B, Cunningham FM. Equine platelets inhibit E. coli growth and can be activated by bacterial lipopolysaccharide and lipoteichoic acid although superoxide anion production does not occur and platelet activation is not associated with enhanced production by neutrophils. Vet Immunol Immunopathol 2013; 152: 209–217.
43. Savcic M, Hauert J & Bachmann F, et al. Clopidogrel loading dose regimens: kinetic profile of pharmacodynamic response in healthy subjects. Semin Thromb Hemost 1999; 25(suppl 2): 15–19.
44. Arora RR, Rai F. Antiplatelet intervention in acute coronary syndrome. Am J Ther 2009; 16: e29–e40.
45. Bonello L, Tantry US & Marcucci R, et al. Consensus and future directions on the definition of high on-treatment platelet reactivity to adenosine diphosphate. J Am Coll Cardiol 2010; 56: 919–933.
46. Chiu FC, Wang TD & Lee JK, et al. Residual platelet reactivity after aspirin and clopidogrel treatment predicts 2-year major cardiovascular events in patients undergoing percutaneous coronary intervention. Eur J Intern Med 2011; 22: 471–477.
47. Tobin WO, Kinsella JA & Coughlan T, et al. High on-treatment platelet reactivity on commonly prescribed antiplatelet agents following transient ischaemic attack or ischaemic stroke: results from the Trinity antiplatelet responsiveness (TRAP) study. Eur J Neurol 2013; 20: 344–352.
48. Brainard BM, Epstein KL & LoBato D, et al. Effects of clopidogrel and aspirin on platelet aggregation, thromboxane production, and serotonin secretion in horses. J Vet Intern Med 2011; 25: 116–122.
49. Ferroni P, Riondino S & Vazzana N, et al. Biomarkers of platelet activation in acute coronary syndromes. Thromb Haemost 2012; 108: 1109–1123.
50. Bowbrick VA, Mikhailidis DP, Stansby G. Value of thromboelastography in the assessment of platelet function. Clin Appl Thromb Hemost 2003; 9: 137–142.
51. Schorer AE, Moldow CF, Rick ME. Interleukin 1 or endotoxin increases the release of von Willebrand factor from human endothelial cells. Br J Haematol 1987; 67: 193–197.
52. Ho-Tin-Noé B, Demers M, Wagner DD. How platelets safeguard vascular integrity. J Thromb Haemost 2011; 9(suppl 1): 56–65.
53. Brainard BM, Epstein KL & LoBato DN, et al. Treatment with aspirin or clopidogrel does not affect equine platelet expression of P selectin or platelet-neutrophil aggregates. Vet Immunol Immunopathol 2012; 149: 119–125.