Platelet activation has been reported for a variety of conditions in dogs, such as IMHA, nephrotic syndrome, parvoviral enteritis, heartworm disease, lymphosarcoma and other malignancies, and sepsis.1–8 In humans, excessive platelet activation promotes thromboembolic events and potentiates inflammatory reactions that can lead to sepsis, septic shock, and MODS.9–13 Once platelets are activated and tethered to the site of endothelial injury, they recruit and activate neutrophils through the action of cell-adhesion receptors, thereby potentiating both the inflammatory and coagulation response.14 In addition, activated platelets release microparticles into the circulation. These microparticles are small, are highly mobile in the microvasculature, have notable procoagulant activity, and may potentially contribute to thrombus formation or disseminated intravascular coagulation.15–18
Systemic inflammation contributes to a hypercoagulable state, likely because of effects attributable to expression of cytokine-mediated tissue factor on the surface of inflammatory cells and on activated endothelium.19 Hypercoagulability is characterized by an increased MA for thromboelastography, which suggests that platelets may play a central role in this procoagulant state.20 The potentially devastating effects of inflammation and hypercoagulability are described in a retrospective study21 of dogs that died or were euthanized because of complications of IMHA. In that study,21 necropsy findings in 28 of 34 (83%) dogs included hepatocellular necrosis and infarcts in the spleen, kidneys, liver, lungs, or heart. In addition, the severity of lesions was correlated with the degree of leukocytosis, which suggests an association between the inflammatory response and hypercoagulability. Given that activated platelets strongly contribute to inflammation, thrombosis, sepsis, MODS, and disseminated intravascular coagulation in a variety of diseases and syndromes, excessive platelet activation may be a consistent finding in critically ill dogs, and therapeutic options to control platelet activation could impact morbidity and mortality rates of these dogs.
Flow cytometry with a variety of fluorescent-labeled antibodies has been used to measure platelet activation in plasma or whole blood of dogs.1–7,22–24 Studies on the use of flow cytometry have focused on specific disease states or conditions, such as Scott syndrome,22 IMHA,7,8 endocardiosis,23 inflammation, or sepsis,4 or on racing Greyhounds after exercise.25 Of these, only the study4 of dogs with SIRS and sepsis had a study population that contained some critically ill dogs. In humans, flow cytometry has been used to detect excessive platelet activation in critically ill patients with coronary artery disease,26 congestive heart failure,27 sepsis,13 and MODS.10,11,28 Similarly, studies16,17 of humans have used flow cytometry to confirm platelet microparticle release in diseases or conditions such as neoplasia, acute coronary syndrome, immune thrombocytopenic purpura, multiple sclerosis, diabetes mellitus, uremia, and systemic lupus erythematosus, which supports a major role of platelets in critical illness.
In contrast to conventional plasma-based coagulation assays, thromboelastography is a whole blood assay that evaluates the dynamic process of coagulation in real time, from the initiation of coagulation through clot formation and fibrinolysis.20 Thromboelastography determines clot strength, speed of clot formation, and rate of clot dissolution through fibrinolysis.29 Thromboelastography has been used in veterinary medicine to confirm hypercoagulability in diseases and conditions, such as IMHA,30,31 immune-mediated thrombocytopenia,32 neoplasia,33 parvovirus infection,5 congenital portosystemic shunts,34 hyperadrenocorticism,35 protein-losing nephropathy,36 protein-losing enteropathy,37,38 and disseminated intravascular coagulation.39 Currently, the authors are not aware of any direct evidence in the veterinary literature that conclusively provides a link between hypercoagulable states identified with thromboelastography and increased platelet activation in naturally occurring disease states. Given the relationship between inflammation and coagulation, it is possible that dogs with SIRS or MODS may have a higher degree of platelet activation and hypercoagulability.
Diagnosis-independent scoring systems (SPI240 and APPLE41) have been developed that can be used to help clinicians predict survival outcome in critically ill dogs. The SPI2 predicts survival outcome in critically ill dogs through multivariate regression of MAP, respiratory rate, creatinine concentration, albumin concentration, age, PCV, and reason for admission (medical vs surgical treatment).40 The APPLE scoring system predicts survival outcome by assigning scores to creatinine concentration, WBC count, albumin concentration, Spo2, total bilirubin concentration, mental status, age, respiratory rate, presence or absence of free fluid (in the abdomen, thorax, or pericardium), and lactate concentration. Increases in APPLE scores correlate with a higher probability of death.41
Determining a positive relationship between platelet activation and hypercoagulability would improve the understanding of hemostatic changes in critically ill dogs and help guide targeted treatment for such dogs. The purpose of the study reported here was to use the SPI2 and APPLE as well as scoring systems for the measurement of SIRS and MODS to compare severity of illness, platelet activation, and hypercoagulability in a population of critically ill dogs. We hypothesized that critically ill dogs would have a greater degree of platelet activation (as measured by use of flow cytometry), compared with results for healthy dogs, and that critically ill dogs would have evidence of hypercoagulability (as assessed by means of thromboelastography). We also hypothesized that the degree of platelet activation would correlate with disease severity, with the greatest platelet activation in dogs with SIRS, MODS, or illness severity scores suggestive of more profound illness.
Materials and Methods
Animals
Critically ill dogs admitted to the Tufts University Cummings School of Veterinary Medicine ICU between September and December 2010 and from March through May 2011 were considered eligible for inclusion. Dogs were enrolled if they had metabolic disease (diabetes mellitus or hyperadrenocorticism), hepatic or renal disease, immune-mediated disease (IMHA or immune-mediated thrombocytopenia), respiratory disease (pneumonia or pleural space disease), cardiovascular disease (valvular disease, cardiomyopathy, or heart failure), sepsis, or severe trauma. Exclusion criteria included age < 1 year, treatment with clopridogrela or aspirin within 7 days before or after admission, treatment with heparin within 30 days before admission, treatment with corticosteroids within 48 hours before or after admission, blood transfusion within 7 days before or after admission, admission to the ICU strictly to enable observation for seizures, and admission to the ICU as a result of a lack of cage space in hospital wards. Twenty-four healthy (control) dogs of various ages and breeds, without a history of underlying illness or injury, were recruited for the study. Control dogs were owned by faculty, staff, and veterinary students. Control dogs were not age-matched to the critically ill dogs. Client consent was obtained for all dogs enrolled in the study. The study was approved by the Tufts Cummings School of Veterinary Medicine Clinical Sciences Review Committee.
Assessment procedures
Within 24 hours after critically ill dogs were admitted to the ICU, they were stratified into groups according to severity of disease on the basis of the SPI2 and APPLE score.40,41 The SPI2 was calculated on the basis of MAP, respiratory rate, serum creatinine concentration, serum albumin concentration, PCV, age, and reason for admission (medical treatment [score, 1] vs surgical treatment [score, 0]). The SPI2 was calculated as logit P and then solved for P, which represented the patient's probability of survival.40 The APPLE score was calculated on the basis of lactate concentration, creatinine concentration, albumin concentration, total bilirubin concentration, total WBC count, Spo2, age, mentation score, fluid score, and respiratory rate. Each of these variables was assigned a numeric score; maximum total potential score was 80, with higher scores representing a higher risk of death.41
Scores for SIRS and MODS were recorded during the first 24 hours of hospitalization. The SIRS score was calculated by assigning a score of 1 for each of the following categories, if present: rectal temperature < 37.51°C or > 39.17°C, heart rate > 140 beats/min, respiratory rate > 40 breaths/min, or WBC count > 16,000 WBCs/μL or < 6,000 WBCs/μL or > 3% band cells.42 Minimum SIRS score was 0, and maximum SIRS score was 4. Dogs with a score ≥ 2 were considered to have SIRS.43 The MODS score was used to classify organ system dysfunction for the coagulation, respiratory, cardiovascular, renal, and hepatic systems. Coagulation dysfunction was considered to have occurred if the platelet count was < 100,000 platelets/μL or if the partial thromboplastin time or activated partial thromboplastin time was > 25% above the upper limit of the respective reference ranges. Renal dysfunction was considered to have occurred if the creatinine concentration was > 2.0 mg/dL with no evidence of prerenal or postrenal azotemia (< 0.5 mg/dL change in subsequent creatinine concentration measured after treatment). Respiratory dysfunction was considered to have occurred if supplemental oxygen or mechanical ventilation was required during hospitalization. Cardiovascular dysfunction was considered to have occurred if a dog had evidence of congestive heart failure, had hypotension that was not responsive to fluid administration, or was hospitalized for treatment of cardiac arrhythmias. Hepatic dysfunction was considered to have occurred if the total bilirubin concentration exceeded 0.5 mg/dL.42
Fluid assessment and mentation scores were assigned within 24 hours after admission for use in calculation of the APPLE score.41 Fluid assessment scores were obtained by use of ultrasonographic evaluation of the thorax and abdomen to determine the presence of peritoneal, pericardial, or pleural effusion.44 A score of 1 was assigned to each compartment that contained free fluid. Thus, a dog with free fluid in the pleural, pericardial, and peritoneal spaces was assigned a score of 3. Mentation scores were assigned as follows: 0, normal mentation; 1, able to stand unassisted and responsive but dull; 2, able to stand with assistance and responsive but dull; 3, unable to stand and responsive but dull; and 4, unable to stand and unresponsive.
The Spo2 was measured for each critically ill dog. The pulse oximeter probe was placed on the nonpigmented part of the upper lip of each dog, and Spo2 was recorded for as long as the appropriate arterial waveform and heart rate were detected.
Survival to discharge from the hospital, total cost of hospitalization, and duration of hospitalization were determined through review of electronic medical records. To determine whether these data points could be used as surrogate indicators for the degree of hypercoagulability in this patient population, they were compared with results for flow cytometry and thromboelastography and other coagulation variables for dogs hospitalized in the ICU.
Vital signs (heart rate, respiratory rate, and rectal temperature) were recorded for both critically ill and healthy control dogs. Blood pressure was measured with an appropriately sized cuff via Doppler ultrasonographic or oscillometric techniques. For each measurement, systolic and diastolic measurements were obtained. The MAP was calculated as follows: MAP = Diastolic blood pressure + 1/3(systolic – diastolic blood pressure).
For most dogs, venipuncture was performed with an 18- or 20-gauge needle to enable collection of 10 mL of blood. When a blood sample was collected through a central venous catheter, the first 6 mL of blood removed from the catheter was discarded prior to collection of the 10-mL sample. Blood collection tubes were filled in the following order: 5.4 mL was placed in 2 sodium citrate tubes, 2 mL was placed in a sodium EDTA tube, and 2 mL was placed in a serum separator tube. The sodium citrate tubes were used for flow cytometry; determination of von Willebrand factor, activated protein C, and antithrombin concentrations; and measurement of partial thromboplastin time and activated partial thromboplastin time. The sodium EDTA tube was used for a CBC, platelet count, and measurement of fibrinogen concentration. The serum separator tube was used for a serum biochemical analysis.
Flow cytometry
Several flow cytometry protocols22,24,45–49 were evaluated during the design phase of the study. Ultimately, a citrated whole blood protocol24 was modified and used to measure platelet activation. Concentrations chosen for all antibodies and agonists were selected on the basis of reported22,24,46,47 consistent and measurable platelet activation.
Antibody buffers were prepared in 50-mL aliquots that contained final concentrations of 10mM HEPES buffer, 0.15M NaCl, and 0.5% bovine serum albumin. Modified HEPES-Tyrode buffer was prepared in 100-mL aliquots that contained final concentrations of 0.14M NaCl, 2.7mM KCl, 1mM MgCl2, 0.25M glucose, 10mM HEPES buffer, 0.5% bovine serum albumin, 0.4mM NaHPO4, and 12mM NaHCO3. Monoclonal mouse anti-human CD61, clone Y2/51, conjugated to allophycocyanin fluorochromeb was used to detect the platelet population. The antibody used to specifically detect activated platelets through P-selectin expression was a monoclonal mouse anti-human CD62-P, clone 1E3, conjugated to phycoerythrin.c Negative isotype control antibodies included normal monoclonal mouse anti-human IgG2a conjugated to phycoerythrinc and normal mouse anti-human IgG1 conjugated to allophycocyanin.b Paraformaldehyde buffer (1%) was prepared by dilution of a stock solution of 30% paraformaldehyde.
Fresh aliquots of all buffers were prepared each week. Antibody buffer was mixed with antibodies and, when appropriate, ADP such that the final concentrations of the anti-CD61, CD62-P, and IgG2a-phycoerythrin antibodies were 50 μg/mL and that of IgG1-allophycocyanin was 100 μg/mL. Final concentration of ADP in the activated reaction buffers was 2, 6, and 10μM. Platelet activation in blood samples from critically ill and control dogs was measured with and without in vitro activation with 2, 6, or 10μM ADP.
Citrated whole blood samples were allowed to sit undisturbed (ie, rest) in a tube rack for 30 minutes at room temperature (approx 23°C). Modified HEPES-Tyrode buffer was used to dilute the blood to a concentration of 1:10. Then, 20 μL of diluted blood was added to tubes that contained 450 μL of modified HEPES-Tyrode buffer at room temperature. An aliquot (20 μL) of the appropriate antibodies with or without ADP buffer was added to each tube. Samples were mixed gently and incubated for 20 minutes at room temperature in the dark, which was followed by the addition of 600 μL of 1% paraformaldehyde fixation solution. Tubes were placed on ice and analyzed within 4 hours.
Flow cytometry was performed with a flow cytometerd calibrated to detect platelets by use of a threshold gate of 800, which effectively excluded all cells not labeled by the CD61 antibody specific to resting or activated platelets. Each run sampled 10,000 events by use of a gate that included platelets and excluded other types of cells on the basis of forward- and side-scatter characteristics. Flow cytometer softwaree measured MFI of the CD62-P–phycoerythrin signal and the percentage of total platelets expressing P-selectin (ie, percentage of cells that recorded CD62-P–phycoerythrin positive events) for cells segregated by use of calibration to detect cells that bound CD62-P–phycoerythrin antibody on activated platelet surfaces. Data analysis was performed by personnel at the University of Massachusetts Flow Cytometry Laboratory by use of the flow cytometer software.f Both MFI and percentage of positive CD62-P events were measured for all sample replicates of each dog.
Thromboelastography
A disposable thromboelastography cup and pin was fitted to the thromboelastograph.g For blood samples collected through an indwelling IV catheter, a disposable thromboelastography cup and pin system impregnated with heparinase was used. An aliquot (20 μL) of CaCl2 was added to overcome effects of citrate-anticoagulated blood. After samples were allowed to sit at room temperature for 30 minutes, 1 mL of citrated whole blood was added to a kaolin hemostasis tube.g Tubes were inverted gently 5 times. Then, 340 μL of kaolin-activated whole blood was added to each cup containing CaCl2, which initiated the reaction. A central pin was lowered into the reaction cup and held in suspension by a torsion wire. The cup was gently oscillated at an angle of 4.75° for 10-second cycles over 30 to 60 minutes. Formation of fibrin between the cup and pin generated tracings from which R (reaction time), α angle, K (clot formation time), MA, and G (global clot strength) were determined.20 On the basis of G, dogs were classified as hypercoagulable (> 7,200 dynes/s), normocoagulable (3,200 to 7,200 dynes/s), or hypocoagulable (< 3,200 dynes/s).39
Statistical analysis
Statistical analysis was performed with a commercial statistical software program.h Values of P ≤ 0.05 were considered significant. The Wilcoxon signed rank test was used to compare median differences of the percentage of positive events and MFI of CD62-P events between resting and ADP-activated platelets within the control and critically ill groups. The Mann-Whitney U test was used to compare results of all laboratory data, flow cytometry data, and thromboelastography data between the control and critically ill groups. The Kruskal-Wallis test was used to compare results of all laboratory data, flow cytometry data, and thromboelastography data between critically ill dogs stratified into groups on the basis of SPI2, APPLE score, SIRS score, MODS score, reason for admission (medical treatment vs surgical treatment), duration of hospitalization, or cost of hospitalization. Dogs were stratified into subgroups on the basis of the interquartile range distribution for each measure. A Spearman correlation test was used to correlate SPI2, APPLE score, cost of hospitalization, and duration of hospitalization with the degree of platelet activation and to correlate thromboelastography results with the degree of platelet activation. Sample size calculations based on results of a 2006 study7 indicated that 80 dogs would provide adequate statistical power (α = 0.05 and power > 80%) to reject the null hypothesis.
Results
Eighty-two critically ill dogs were enrolled in the study. The 82 dogs represented 39 breeds, including 16 mixed-breed dogs, 7 Golden Retrievers, 4 Standard Poodles, 4 German Shepherd Dogs, 3 Doberman Pinschers, 3 Pugs, 3 Mastiffs, 3 English Bulldogs, 2 Dachshunds, 2 Boxers, 2 Beagles, 2 Labrador Retrievers, 2 English Sheepdogs, 2 German Shorthaired Pointers, 2 West Highland White Terriers, and 2 American Pit Bull Terriers; 23 other breeds were represented by 1 dog each. There were 37 spayed females, 4 sexually intact females, 32 castrated males, and 9 sexually intact males. Mean age was 7.86 years. The 24 control dogs represented 12 breeds, including 8 mixed-breed dogs, 4 Golden Retrievers, 2 Labrador Retrievers, and 2 American Pit Bull Terriers; 8 breeds were represented by 1 dog each. There were 12 spayed females, 3 sexually intact females, 7 castrated males, and 2 sexually intact males. Mean age was 5.54 years. There was no significant difference in age between healthy control and critically ill dogs.
Of the 82 critically ill dogs, 53 were hospitalized in the ICU for medical treatment and 29 were hospitalized for surgical treatment. Of the 29 dogs hospitalized for surgical conditions, 10 did not undergo surgery because of client financial constraints or a poor prognosis. Overall, 21 dogs were hospitalized because of known or suspected neoplasia that resulted in critical illness. Within the medical treatment group, cardiac disease (n = 11), renal disease (6), neurologic disease (6), hepatic disease (5), pneumonia (5), pulmonary hypertension (3), immune-mediated disease (2), hypoadrenocorticism (2), and pancreatitis (2) were the most common causes for hospitalization. Two dogs within the medical treatment group had lethargy and fever of unknown origin for which no cause was determined. Within the surgical treatment group, neoplasia (n = 6), sepsis (5), trauma (4), nontraumatic hemoperitoneum (4), gastric-dilatation or gastric dilatation-volvulus syndrome (3), and lung lobe torsion (2) were the most common reasons for surgery.
Overall, 61 (74.4%) dogs survived to discharge from the hospital. Seventeen (20.7%) dogs were euthanized during the course of hospitalization because of a poor prognosis or acute decline in condition during hospitalization. No dogs were euthanized strictly because of the cost of care. Four (4.9%) dogs died during hospitalization as a result of severity of illness. Two dogs that survived to discharge were euthanized within 2 weeks after leaving the ICU. Dogs were hospitalized in the ICU for a median of 3 days (range, 1 to 11 days) and at a median cost of $2,751 (range, $357 to $12,243).
Dogs were scored for SIRS and MODS on a scale of 0 to 4, as described previously. Forty-one dogs had a SIRS score ≥ 2 (score of 2 for 30 dogs and score of 3 for 11 dogs); thus, SIRS was diagnosed. Eight dogs were excluded from SIRS scoring because they lacked a laboratory WBC count. The most common diseases or conditions associated with SIRS included confirmed or suspected sepsis (n = 11), neoplasia (10), congestive heart failure (3), acute kidney injury (3), dilated cardiomyopathy (2), hypoadrenocorticism (2), and liver failure (2). Most dogs did not have evidence of MODS. Thirty-three dogs had no evidence of organ dysfunction, and 32 dogs had dysfunction in only 1 organ system. Eleven dogs had dysfunction in 2 or 3 organ systems; thus MODS was diagnosed. Diseases or conditions for the 11 dogs with MODS included pericardial effusion with or without a visible cardiac mass (n = 3), acute kidney injury (2), metastatic neoplasia (2), IMHA (1), congestive heart failure (1), pulmonary hypertension (1), and pyothorax (1).
Macrothromboembolic disease was detected in 3 dogs during surgery, diagnostic imaging, or necropsy. One dog had a large mass in the left atrium and fully occlusive thrombus at the bifurcation of the abdominal aorta and femoral arteries and was euthanized shortly after emergency admission. This dog had the greatest degree of platelet activation in a resting sample as measured with flow cytometry. The other 2 dogs had visible thrombi that were detected during surgery: one dog had small thrombi throughout the pleural cavity attributable to pyothorax, and the other had multiple thrombi throughout the spleen secondary to hemangiosarcoma.
Thromboelastography classified dogs as hypercoagulable (n = 56), normocoagulable (19), and hypocoagulable (7), as assessed on the basis of G. The hypercoagulable group consisted of dogs with sepsis (n = 9), acute kidney injury (5), pulmonary hypertension (3), and pancreatitis (2), whereas none of the normocoagulable or hypocoagulable dogs had these diseases or conditions. Pneumonia, trauma, and congestive heart failure were represented most frequently among hypercoagulable and normocoagulable dogs, whereas cardiac disease, nontraumatic hemoperitoneum, and neoplasia were detected in all dogs regardless of coagulation status. These groups were not of sufficient size to allow statistical comparisons.
Critically ill dogs had significant increases in MA, α angle, and G and significantly lower K, compared with values for control dogs. In addition, R was prolonged in critically ill dogs, but not significantly, compared with the result for control dogs (Table 1). There were no differences in thromboelastographic variables among critically ill dogs when subclassified on the basis of SIRS scores, MODS scores, reason for admission (medical treatment vs surgical treatment), or cost or duration of hospitalization.
Median (range) thromboelastography values for critically ill dogs and healthy control dogs.
| Critically ill dogs | Control dogs | ||||
|---|---|---|---|---|---|
| Variable | n | Median (range) | n | Median (range) | P value* |
| K (min) | 80 | 1.85 (0.80–17.90) | 24 | 2.60 (1.80–4.80) | < 0.001 |
| R (min) | 82 | 5.50 (2.40–9.90) | 24 | 4.85 (2.70–10.30) | 0.164 |
| α angle (°) | 82 | 63.95 (19.30–82.50) | 24 | 57.85 (38.70–65.30) | 0.002 |
| MA (mm) | 82 | 64.85 (12.50–86.90) | 24 | 54.05 (47.90–61.40) | < 0.001 |
| G (dynes/s) | 82 | 9,228.6 (711.2–33,140.6) | 24 | 5,888.1 (0–7,966.0) | < 0.001 |
Values were considered significant at P ≤ 0.05.
G = Global clot strength. K = Clot formation time. R = Reaction time.
Critically ill dogs had significant increases in platelet reactivity to ADP, as measured by both the percentage of positive events and MFI for the CD62-P (P-selectin) antibody signal, compared with results for control dogs. There were no significant differences between the proportion of circulating activated platelets in critically ill dogs, compared with that of healthy control dogs, in samples that were not stimulated with ADP, although critically ill dogs had both a higher percentage of positive events and greater MFI for the P-selectin signal. Critically ill dogs had significant increases in platelet activation when whole blood samples were stimulated with 2, 6, or 10μM ADP for measurements of MFI or percentage of positive events (Table 2). There were no differences in P-selectin expression in resting samples or samples stimulated with ADP among critically ill dogs classified on the basis of SIRS scores, MODS scores, or reason for admission (medical treatment vs surgical treatment).
Median (range) flow cytometry values for blood samples from critically ill dogs and healthy control dogs.
| Positive events (%) | MFI (arbitrary units) | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Critically ill dogs | Control dogs | Critically ill dogs | Control dogs | |||||||
| Variable | n | Median (range) | n | Median (range) | P value* | n | Median (range) | n | Median (range) | P value* |
| Resting† | 78 | 0.57 (0.0–7.65) | 24 | 0.30 (0.0–5.38) | 0.230 | 78 | 306 (121–975,000) | 24 | 523 (130–5,654) | 0.723 |
| 2μM ADP | 76 | 5.76 (0–81.10) | 24 | 3.20 (0–14.70) | < 0.001 | 76 | 916 (123–70,700) | 24 | 508 (0–21,500) | < 0.001 |
| 6μM ADP | 78 | 21.05 (0.01–91.30) | 24 | 15.60 (0.0–42.5) | 0.028 | 78 | 1,818 (110–23,800) | 24 | 782 (258–22,700) | < 0.001 |
| 10μM ADP | 82 | 51.90 (0–91.20) | 17 | 28.70 (0.34–50.90) | 0.263 | 72 | 2,256 (187–31,700) | 21 | 1,080 (426–3,733) | < 0.001 |
Represents unstimulated whole blood.
See Table 1 for remainder of key.
Flow cytometry data were visually examined (Figure 1). No significant correlations were identified between MA or G and any measure of platelet activation assessed by use of flow cytometry.
Representative flow cytometry output for a critically ill dog. A—Scatterplot of a resting (nonstimulated) whole blood sample. On the basis of forward-scatter (FSC-H) and side-scatter (SSC-H) properties, the majority of platelets are in gate P1. B—Same scatterplot as panel A after the whole blood sample was stimulated with 10μM ADP. The MFI for CD62-P–phycoerythrin is determined from platelets within gate P1. C—Histogram for a resting whole blood sample that indicates counts of activated platelets detected at the FL2-H sensor, which detects the phycoerythrin fluorochrome tagged to the CD62-P antibody that binds to activated platelets. Notice the gated area in which activated platelets would be detected (M1), which is detecting 7.1% of total activated platelets. D—Same histogram as in panel C after the whole blood sample was stimulated with 10μM ADP. Notice the profound increase in the percentage of activated platelets (72.9%) within the M1 gate.
Citation: American Journal of Veterinary Research 76, 4; 10.2460/ajvr.76.4.328
Expression of P-selectin was compared between control dogs and critically ill dogs identified on the basis of G as hypercoagulable (n = 52), normocoagulable (20), or hypocoagulable (7). Dogs identified as hypercoagulable and normocoagulable had significant increases in percentage of positive events when samples were stimulated with 2, 6, or 10μM ADP but did not differ in percentage of positive events or MFI for resting (unstimulated) samples. There were no significant differences in percentage of positive events or MFI between these hypercoagulable and normocoagulable dogs. No statistical comparisons were performed for hypocoagulable dogs because of the small sample size of this group.
When critically ill dogs were stratified into subgroups on the basis of SPI2, APPLE score, SIRS score, or MODS score or by cost or duration of hospitalization, dogs stratified by APPLE score had differences in the percentage of positive events for blood stimulated with 10μM ADP. Correlation analysis revealed a weak but significant negative relationship (R = −0.347; P = 0.012) between APPLE score and platelet activity for blood stimulated with 10μM ADP. When stratified into groups on the basis of interquartile range for APPLE scores, dogs with the lowest APPLE scores (< 22) had significantly greater platelet activation for blood stimulated with 10μM AD, compared with results for all other subgroups. No other flow cytometry or thromboelastography variables were correlated with any measure of illness severity.
Coagulation variables, including fibrinogen concentration, partial thromboplastin time, activated partial thromboplastin time, von Willebrand factor concentration, antithrombin concentration, and activated protein C concentration, were compared between critically ill dogs and healthy control dogs. Critically ill dogs had a significantly higher fibrinogen concentration, prolonged partial thromboplastin time, and activated partial thromboplastin time and a significantly decreased percentage of measurable antithrombin, compared with results for control dogs (Table 3). No differences were found between critically ill dogs stratified on the basis of SPI2, APPLE score, SIRS score, or MODS score or by cost or duration of hospitalization. No correlations were found between coagulation variables and platelet activation or hypercoagulability.
Median (range) results of coagulation panel testing for critically ill dogs and healthy control dogs.
| Critically ill dogs | Control dogs | ||||
|---|---|---|---|---|---|
| Variable | n | Median (range) | n | Median (range) | P value* |
| Fibrinogen (mg/dL) | 36 | 374 (60–773) | 13 | 164 (126–220) | < 0.001 |
| Prothrombin time (s) | 37 | 8.2 (6.3–11.8) | 14 | 7.5 (6.5–8.7) | 0.029 |
| Activated partial thromboplastin time (s) | 36 | 15.4 (9.9–31.5) | 14 | 12.7 (11.0–15.1) | 0.007 |
| Antithrombin (%) | 37 | 94.6 (44.5–138.0) | 17 | 124.0 (68.3–139.0) | 0.003 |
| von Willebrand factor (%) | 36 | 76.4 (9.2–139.0) | 14 | 90.4 (72.0–132.0) | 0.070 |
| Activated protein C (%) | 33 | 100.0 (53.4–183.0) | 7 | 91.4 (36.6–121.0) | 0.293 |
See Table 1 for key.
Discussion
The present study provided evidence that platelets in healthy dogs are inherently more resistant to ADP-dependent activation than are platelets from critically ill dogs. This evidence supports the notion that measuring platelet reactivity to exogenous agonists may provide a more reliable assessment of the overall functional status of circulating platelets in sick dogs than analysis of platelet activation in nonstimulated samples. Platelets that are activated in vivo may be rapidly removed from the circulation and may not reflect the true activation status of circulating platelets in sick animals.50 Activated platelets can rapidly lose surface expression of P-selectin but continue to circulate and function in platelet aggregation, fibrinogen binding, provision of a procoagulant surface, and shedding of procoagulant microparticles.51 Therefore, addition of a platelet agonist in vitro may be useful in determining platelet hyperreactivity and may be considered a useful indicator of the capacity of circulating platelets to contribute to the prothrombotic state often encountered in patient populations such as dogs hospitalized in an ICU.
P-selectin was used as the marker of platelet activation in this population of dogs. The P-selectin expression for healthy control dogs, as measured by percentage of positive CD62-P events, was similar to that described in veterinary studies8,22,24,46 that involved use of a whole blood flow cytometry protocol. However, results for maximal median P-selectin expression after activation with 10μM ADP differed from those of 2 other studies24,46 in which investigators used ADP as a platelet agonist for flow cytometry. Within the study population reported here, there was some variation among subjects with regard to robustness of ADP-dependent platelet activation. The disparity seen with ADP-dependent platelet activation in dogs is likely multifactorial and may be attributable to variation within individuals and study populations or subtle differences in protocols among laboratories. Historically, ADP has been considered a weaker platelet agonist in aggregation evaluations as a result of variable individual responses and inconsistency when used in whole blood assays.7,50 Agonists such as collagen, thrombin, platelet activating factor, and phorbol myristate acetate are viewed as stronger platelet activators,50 and some of them have been successfully used as agonists in other veterinary studies.7,8,24,25,46,47 Future studies that involve more potent agonists may be useful for examining platelet activation in disease states.
Most dogs admitted to the ICU in the present study were hypercoagulable as determined on the basis of G, K, MA, and α angle. These results are similar to findings from previous studies in which investigator found a hypercoagulable state in dogs with severe diseases and conditions such as IMHA,30,31 immune-mediated thrombocytopenia,32 protein-losing enteropathy,37 protein-losing nephropathy,36 metastatic neoplasia,33 and disseminated intravascular coagulation.39 However, no significant correlation was detected between the degree of ADP-stimulated platelet activation and the thromboelastography variables MA or G. It is possible that ADP, as a weak agonist, did not provide a strong enough stimulus to enable effective measurement of platelet hyperreactivity when used at lower concentrations. The MA may also be affected by factors other than platelet function, such as fibrinogen concentration. In the present study, fibrinogen concentration was significantly higher in critically ill dogs and may have contributed to clot strength and therefore the increased MA detected in this group. Finally, anemia can result in an artifactual increase in MA. Some dogs in this study developed anemia, which may have contributed to the lack of correlation between thromboelastography variables (eg, MA) and ADP-dependent platelet activation.
Neither the SPI2 nor APPLE scoring systems were predictive of the proportion of activated platelets or hypercoagulability in dogs in the study reported here. Several factors may explain this finding. First, a minimum SPI2 or APPLE score was not required for study inclusion. Incorporation of a minimum APPLE or SIRS score into the inclusion criteria for enrollment would have narrowed the population to sicker dogs and may have increased the likelihood of detecting changes in hemostasis or platelet activation in the patient population. Additionally, at least 20% of the patient population did not qualify for scoring because they did not have results for the appropriate laboratory tests within 24 hours after admission. As a result, the relationship between these scores and platelet activation did not reflect the entire population. Finally, it is possible that both the SIRS and APPLE scoring systems were inappropriate for use in predicting hypercoagulability in critically ill dogs. Both systems were developed for use in predicting survival to discharge on a population basis and were not intended for use in predicting outcome for individual animals or predicting the amount of hypercoagulability or platelet activation. Future studies could focus on developing a diagnosis-specific scoring system for predicting hypercoagulability and outcome, much as the International Renal Interest Society scoring system is used to assess renal disease outcome.52 To the authors’ knowledge, there currently are no scoring systems in human or veterinary medicine designed specifically for assessing platelet activation or hypercoagulability with clinical outcome.
In the present study, we did not detect a relationship between cost or duration of hospitalization and any measures of platelet activation or hypercoagulability as measured by use of flow cytometry or thromboelastography. Similarly, cost or duration of hospitalization did not correlate with any of the 4 calculable measures of illness severity (SIRS score, MODS score, SPI2, or APPLE score).
The increased fibrinogen concentration, decreased antithrombin concentration, and prolonged partial thromboplastin and activated partial thromboplastin times for critically ill dogs in the present study were within laboratory reference ranges and therefore probably were not clinically relevant. The SIRS and MODS scores also had no relationship to the proportion of activated platelets or degree of hypercoagulability as measured by use of flow cytometry or thromboelastography. Only 41 (50%) dogs of the study population had SIRS, whereas 11 (13%) dogs had MODS. Future studies could focus on screening dogs by use of SIRS or MODS scoring before enrollment. By selecting for dogs with SIRS or MODS, a uniform population of sicker dogs could be evaluated.
The exclusion of dogs that recently received corticosteroids, NSAIDs, or blood products was a major limiting factor for this study. Many dogs with immune-mediated diseases such as IMHA, immune-mediated thrombocytopenia, or inflammatory bowel disease were excluded because of prior corticosteroid administration, despite evidence that they were severely ill. A number of dogs with severe trauma were excluded because the dogs had received NSAIDs prior to referral. Also, severely and acutely hypovolemic dogs that required blood transfusions at the time of admission to our emergency department often received blood products before samples could be collected for flow cytometry or thromboelastography. As a result, the opportunity was lost to include these critically ill dogs that were subsequently admitted to the ICU during the study period.
Although P-selectin is a robust marker for platelet activation, its use as a singular marker in the present study may have limited the ability to detect the true spectrum of activated platelets in critically ill dogs. It has been suggested that P-selectin–positive platelets may be buried in platelet-leukocyte aggregates and may not be detectable with standard flow cytometry assays that use CD62-P as the only marker for platelet activation.46 Therefore, in vivo increases in platelet activation for nonstimulated blood samples cannot be ruled out despite the lack of significant effects for P-selectin in the study reported here. Platelet-leukocyte aggregates are considered a sensitive marker of platelet activation in humans and have been used in several veterinary studies8,46 to measure platelet activation in dogs. Platelet microparticles are readily shed from the lipid membrane of activated platelets and contain many of the surface elements necessary for leukocyte activation, platelet activation, and support of coagulation.16 Platelet microparticles are frequently measured in humans16 and have been investigated in at least 3 veterinary studies8,25,46 regarding platelet activation. Additionally, mean platelet concentration, an automated measure of platelet refractive index, has been indicated as a possible method of measuring platelet activation.50 As platelets activate and degranulate, mean platelet concentration should theoretically decrease. Mean platelet concentration has been associated with platelet activation during nonseptic inflammation in dogs.47 Future studies should focus on the use of these markers in combination with P-selectin expression to further define the extent of platelet activation in critically ill dogs.
In the present study, a large population of critically ill dogs had increased platelet activation in blood samples in response to stimulation with ADP as measured by P-selectin expression. The same population had evidence of hypercoagulability as measured by use of thromboelastography. Analysis of these results suggested that critically ill dogs have hyperreactive platelets, which may in part contribute to a prothrombotic state. Overall, the severity of illness scoring systems tested (including APPLE score, SPI2, SIRS score, and MODS score) or cost or duration of hospitalization did not strongly correlate with the proportion of activated platelets or degree of hypercoagulability in this patient population.
Acknowledgments
This manuscript represents a portion of a thesis submitted by Dr. Majoy to the Tufts University Department of Biomedical Sciences as partial fulfillment of the requirements for a Master of Science degree.
Supported by the Companion Animal Fund at Tufts Cummings School of Veterinary Medicine and by the Tufts Coagulation Laboratory.
Presented in abstract form at the American College of Veterinary Internal Medicine Forum, New Orleans, June 2012.
The authors thank Dawn Meola for technical assistance.
ABBREVIATIONS
| APPLE | Acute patient physiology and laboratory evaluation |
| ICU | Intensive care unit |
| IMHA | Immune-mediated hemolytic anemia |
| MA | Maximum amplitude |
| MAP | Mean arterial blood pressure |
| MFI | Median fluorescence intensity |
| MODS | Multiple organ dysfunction syndrome |
| SIRS | Systemic inflammatory response syndrome |
| SPI2 | Revised survival prediction index score |
| Spo2 | Oxygen saturation as measured by pulse oximetry |
Footnotes
Plavix, Bristol-Myers Squibb, New York, NY.
Dako, Glostrup, Denmark.
Santa Cruz Biotechnology, Santa Cruz, Calif.
Accuri C6 flow cytometer system, BD Accuri Cytometers, Grand Rapids, Mich.
C-Flow, BD Accuri Cytometers, Grand Rapids, Mich.
FloJo flow cytometry analysis software, version 7.6.4, FlowJo LLC, Ashland, Ore.
TEG, Haemoscope, Niles, Ill.
SPSS, version 17.0, SPSS, Chicago, Ill.
References
1. Boudreaux MK, Dillon AR, Spano JS. Enhanced platelet reactivity in heartworm-infected dogs. Am J Vet Res 1989; 50: 1544–1547.
2. Green RA, Russo EA, Greene RT, et al. Hypoalbuminemia related platelet hypersensitivity in two dogs with nephrotic syndrome. J Am Vet Med Assoc 1985; 186: 485–488.
3. McNiel EA, Ogilvie GK, Fettman MJ, et al. Platelet hyperfunction in dogs with malignancies. J Vet Intern Med 1997; 11: 178–182.
4. Moritz A, Walcheck BK, Weiss DJ. Evaluation of flow cytometric and automated methods for detection of activated platelets in dogs with inflammatory disease. Am J Vet Res 2005; 66: 325–329.
5. Otto CM, Rieser TM, Brooks MB, et al. Evidence of hypercoagulability in dogs with parvoviral enteritis. J Am Vet Med Assoc 2000; 217: 1500–1504.
6. Thomas JS, Rogers KS. Platelet aggregation and adenosine triphosphate secretion in dogs with untreated multicentric lymphoma. J Vet Intern Med 1999; 13: 319–322.
7. Weiss DJ, Brazzell JL. Detection of activated platelets in dogs with primary immune-mediated hemolytic anemia. J Vet Intern Med 2006; 20: 682–686.
8. Ridyard AE, Shaw DJ, Milne EM. Evaluation of platelet activation in canine immune-mediated haemolytic anaemia. J Small Anim Pract 2010; 51: 296–304.
9. Amaral A, Opal SM, Vincent JL. Coagulation in sepsis. Intensive Care Med 2004; 30: 1032–1040.
10. Gawaz M, Dickfield T, Bogner C, et al. Platelet function in septic multiple organ dysfunction syndrome. Intensive Care Med 1997; 23: 379–385.
11. Russwurm S, Vickers J, Meier-Hellmann A, et al. Platelet and leukocyte activation correlate with the severity of septic organ dysfunction. Shock 2002; 17: 263–268.
12. Vincent JL, Yagushi A, Pradier O. Platelet function in sepsis. Crit Care Med 2002; 30: S313–S317.
13. Yaguchi A, Lobo FLM, Vincent JL, et al. Platelet function in sepsis. J Thromb Haemost 2004; 2: 2096–2102.
14. Evangelista V, Smyth SS. Interaction between platelets, leukocytes, and the endothelium. In: Michelson A, ed. Platelets. 2nd ed. San Diego: Elsevier Science, 2002; 295–312.
15. Zwicker JI. Tissue factor-bearing microparticles and cancer. Semin Thromb Hemost 2008; 34: 195–198.
16. Nomura S. Function and clinical significance of platelet-derived microparticles. Int J Hematol 2001; 74: 397–404.
17. Ardoin SP, Shanahan JC, Pisetsky DS. The role of microparticles in inflammation and thrombosis. Scand J Immunol 2007; 66: 159–165.
18. Langer F, Spath B, Haubold K, et al. Tissue factor procoagulant activity of plasma microparticles in patients with cancer-associated disseminated intravascular coagulation. Ann Hematol 2008; 87: 451–457.
19. Smith SA. The cell-based model of coagulation. J Vet Emerg Crit Care (San Antonio) 2009; 19: 3–10.
20. Donahue SM, Otto CM. Thromboelastography: a tool for measuring hypercoagulability, hypocoagulability, and fibrinolysis. J Vet Emerg Crit Care (San Antonio) 2005; 15: 9–16.
21. McManus PM, Craig LE. Correlation between leukocytosis and necropsy findings in dogs with immune-mediated hemolytic anemia: 34 cases (1994–1999). J Am Vet Med Assoc 2001; 218: 1308–1313.
22. Brooks MB, Randolph J, Warner K, et al. Evaluation of platelet function screening tests to detect platelet procoagulant deficiency in dogs with Scott syndrome. Vet Clin Pathol 2009; 38: 306–315.
23. Tarnow I, Kristensen AT, Olsen LH, et al. Dogs with heart diseases causing turbulent high-flow velocity blood flow have changes in platelet function and von Willebrand factor multimer distribution. J Vet Intern Med 2005; 19: 515–522.
24. Tarnow I, Kristensen AT, Krogh AKR, et al. Effects of physiologic agonists on canine whole blood flow cytometry assays of leukocyte-platelet aggregation and platelet activation. Vet Immunol Immunopathol 2008; 123: 345–352.
25. Moritz A, Walcheck BK, Deye J, et al. Effects of short-term racing activity on platelet and neutrophil activation in dogs. Am J Vet Res 2003; 64: 855–859.
26. Linden MD, Frelinger AL, Barnard MR, et al. Application of flow cytometry to platelet disorders. Semin Thromb Hemost 2004; 30: 501–511.
27. Chung I, Choudhury A, Lip GYH. Platelet activation in acute, decompensated congestive heart failure. Thromb Res 2007; 120: 709–713.
28. Soriano AO, Wenche J, Chirinos JA, et al. Levels of endothelial and platelet microparticles and their interactions with leukocytes negatively correlate with organ dysfunction and predict mortality in severe sepsis. Crit Care Med 2005; 33: 2540–2546.
29. Mallett SV, Cox DJ. Thromboelastography. Br J Anaesth 1992; 69: 307–313.
30. Fenty RK, de Laforcade AM, Shaw SE, et al. Identification of hypercoagulability in dogs with primary immune-mediated hemolytic anemia by means of thromboelastography. J Am Vet Med Assoc 2011; 238: 463–467.
31. Sinnott VB, Otto CM. Use of thromboelastography in dogs with immune-mediated hemolytic anemia: 39 cases (2000–2008). J Vet Emerg Crit Care (San Antonio) 2009; 19: 484–488.
32. O'Marra SK, Shaw SP, de Laforcade AM. Investigating hypercoagulability during treatment for immune-mediated thrombocytopenia: a pilot study. J Vet Emerg Crit Care (San Antonio) 2011; 22: 128–130.
33. Kristensen AT, Wiinberg B, Jessen LR, et al. Evaluation of human recombinant tissue-factor activated in 49 dogs with neoplasia. J Vet Intern Med 2008; 22: 140–147.
34. Kelley D, Lester C, de Laforcade A, et al. Thromboelastographic evaluation of dogs with congenital portosystemic shunts. J Vet Intern Med 2013; 27: 1262–1267.
35. Rose L, Dunn ME, Bedard C. Effect of canine hyperadrenocorticism on coagulation parameters. J Vet Intern Med 2013; 27: 207–211.
36. Donahue SM, Brooks M, Otto CM. Examination of hemostatic parameters to detect hypercoagulability in dogs with severe protein-losing nephropathy. J Vet Emerg Crit Care (San Antonio) 2011; 21: 346–355.
37. Goodwin LV, Goggs R, Chan DL, et al. Hypercoagulability in dogs with protein-losing enteropathy. J Vet Intern Med 2011; 25: 273–277.
38. Lennon EM, Hanel RM, Walker JM, et al. Hypercoagulability in dogs with protein-losing enteropathy as assessed by thromboelastography. J Vet Intern Med 2013; 27: 462–468.
39. Wiinberg B, Jensen PI, Johansson E, et al. Thromboelastographic evaluation of hemostatic function in dogs with disseminated intravascular coagulation. J Vet Intern Med 2008; 22: 357–365.
40. King LG, Wohl JS, Manning AM, et al. Evaluation of the survival prediction index as a model of risk stratification for clinical research in dogs admitted to intensive care units at four locations. Am J Vet Res 2001; 62: 948–954.
41. Hayes G, Mathews K, Doig G, et al. The acute patient physiologic and laboratory evaluation (APPLE) score: a severity of illness stratification system for hospitalized dogs. J Vet Intern Med 2010; 24: 1034–1047.
42. Kenney EM, Rozanski EA, Rush JE, et al. Association between outcome and organ system dysfunction in dogs with sepsis: 114 cases (2003–2007). J Am Vet Med Assoc 2010; 236: 83–87.
43. De Laforcade A. Systemic inflammatory response syndrome. In: Silverstein DC, Hopper K, eds. Small animal critical care medicine. St Louis: Saunders Elsevier, 2009; 46–49.
44. Lisciandro GR. Abdominal and thoracic focused assessment with sonography for trauma, triage, and monitoring in small animals. J Vet Emerg Crit Care (San Antonio) 2011; 21: 104–122.
45. Dircks BH, Scuberth H-J, Minschke R. Underlying diseases and clinicopathologic variables of thrombocytopenic dogs with and without platelet-bound antibodies detected by use of a flow cytometric assay: 83 cases (2004–2006). J Am Vet Med Assoc 2009; 235: 960–966.
46. Wills TB, Wardrop J, Meyers KM. Detection of activated platelets in canine blood by use of flow cytometry. Am J Vet Res 2006; 67: 56–63.
47. Moritz A, Walcheck BK, Weiss DJ. Flow cytometric detection of activated platelets in the dog. Vet Clin Pathol 2003; 32: 6–12.
48. Matzdorff A. Platelet function tests and flow cytometry to monitor antiplatelet therapy. Semin Thromb Hemost 2005; 31: 393–399.
49. Weiss DJ. Application of flow cytometric techniques to veterinary clinical hematology. Vet Clin Pathol 2002; 31: 72–82.
50. Wiinberg B, Jessen LR, Tarnow I, et al. Diagnosis and treatment of platelet hyperreactivity in relation to thrombosis in dogs and cats. J Vet Emerg Crit Care (San Antonio) 2012; 22: 42–58.
51. Michelson AD, Barnard MR, Hechtman HB, et al. In vivo tracking of platelets: circulating degranulated platelets rapidly lose surface P-selectin but continue to circulate and function. Proc Natl Acad Sci U S A 1996; 93: 11877–11882.
52. Hayes G, Mathews K, Kruth S, et al. Illness severity scores in veterinary medicine: what can we learn? J Vet Intern Med 2010; 24: 457–466.
