Thromboembolic disease is a common and potentially fatal complication of IMHA in dogs.1–4 Corticosteroids are routinely used in the treatment of IMHA; however, hyperadrenocorticism and corticosteroid administration have been identified as risk factors for TED in several retrospective studies2,5–7 of dogs with pulmonary, cranial vena cava, and aortic thrombosis. This raises a concern that corticosteroid use in IMHA may promote TED.
Heparin has been used for many years in efforts to prevent TED in canine patients with IMHA. More recently, anticoagulant treatment with ASA at a low dosage (0.5 mg/kg/d) has been reported to improve survival rates in dogs with IMHA.3 The effect of this low dosage of ASA on platelet function in dogs is controversial. In 1 study,a treatment with 0.5 mg of ASA/kg/d did not affect platelet function assessed by means of optical aggregometry, whereas another studyb in which whole blood aggregometry was used demonstrated decreased platelet aggregation following a similar ASA treatment regimen. Both studies were performed on healthy dogs without the administration of corticosteroids, so it is unknown whether ASA would provide protection against hypercoagulability promoted by corticosteroid treatment. Differences in aggregometry techniques and platelet agonists used in platelet function testing may have accounted for the discrepancy in results between studies.
Thromboelastography is a method of assessing the speed and strength of fibrin clot formation in vitro. Whole blood is added to a cup, into which a pin attached to a torsion wire is inserted. Developed in 1948, TEG is not a new technology, but recent introduction of computerized methods and dedicated software has made it a useful point-of-care test.8 Because the assay is performed with whole blood, it is thought to better represent in vivo hemostasis, compared with traditional plasma-based methods, such as PT and PTT. Thrombo-elastography can also be used to detect more rapid clot formation and increased clot strength, which are suggestive of a hypercoagulable state.8 Hypercoagulability as defined by use of TEG variables has been previously described in dogs with IMHA.9,c,d However, anemia has been shown to increase MA, and the contribution of Hct to the hypercoagulable TEG tracings in dogs of that study was unknown.9–11,e Thromboelastography has also been used to monitor anticoagulant treatment in humans.12 Its usefulness in monitoring the effects of anticoagulant medication remains unproven in veterinary medicine, and results of a recent study13 suggested that it may be too sensitive to the effects of unfractionated heparin to be a useful tool in this application.
The purpose of the study reported here was to evaluate the effect of an immunosuppressive dosage of prednisone (approx 2 mg/kg/d), ASA at a low dosage (approx 0.5 mg/kg/d), and the combination of these 2 drugs on TEG variables in healthy dogs. Our hypotheses were that prednisone would cause samples evaluated by means of TEG to appear relatively hypercoagulable, compared with values obtained at baseline (ie, immediately before treatment) and in untreated control dogs, and that a low dosage of ASA would have no effect on TEG variables.
Materials and Methods
Animals—Two purpose-bred 1.5-year-old Beagles (dogs 1 and 2; both sexually intact males) were evaluated in a preliminary experiment to determine the duration of effect of prednisone on TEG variables. Sixteen purpose-bred mixed-breed hounds (all males; 15 sexually intact and 1 neutered) were enrolled in the main study. These did not include the 2 dogs evaluated in preliminary experiments. Median age of the dogs was 2.6 years (range, 1 to 7 years), and median weight was 26.7 kg (range, 17.7 to 37.3 kg). All dogs were deemed healthy on the basis of history and results of a physical examination, CBC, serum biochemical analysis, and coagulation profile (PT, PTT, and plasma fibrinogen concentration). All procedures met the guidelines set by the Canadian Council on Animal Care.14,15 The study was approved by the Animal Care Committee at the University of Guelph and was performed in accordance with the Animals for Research Act (Ontario 1980).
Study design—The 2 Beagles in the preliminary experiment were administered prednisonef (2 mg/kg, PO, q 24 h for 7 days). Initial blood samples were collected as described for the main study on days 0 (immediately before the first treatment; baseline value), 2, 4, and 6. Blood samples were subsequently collected from these dogs every 7 days for 4 weeks to complete the preliminary experiment.
For the main study, a power calculation indicated that 17 dogs/group would be necessary to ensure a power of 0.80 to detect a 10% difference in MA between groups. A convenience sample size of 16 dogs was used. Dogs were randomly assigned into 3 treatment groups and 1 control group (4 dogs/group) by use of a computerized random-number generator.16 One group received prednisonef,g at a median dosage of 2.07 mg/kg, PO, every 24 hours (range, 1.94 to 2.11 mg/kg) on days 0 through 6 (prednisone group). Another group received ASAh at a median dosage of 0.51 mg/kg, PO, every 24 hours (range, 0.46 to 0.53 mg/kg) on days 0 through 6 (ASA group). Because the ASA dose was small, the drug was compounded by weighti into 2- and 5-mg doses in gel capsulesj to permit accurate dose administration. A third group received prednisone and ASA simultaneously at the described dosages (drug combination group) on days 0 through 6. The remaining group (control group) received no treatment.
Blood samples for CBC, serum biochemical analysis, a coagulation profile, and TEG were obtained from every dog on day 0 immediately before the first treatment (ie, baseline). Thereafter, samples for platelet count and TEG were obtained on days 2, 4, and 6. Food was not withheld prior to collection of samples. Because the groups were studied concurrently but only 2 blood samples could be analyzed simultaneously by use of the TEG instrument, the samples were obtained at fixed intervals throughout the day. Each individual dog had a blood sample collected, then was administered its treatment (if applicable) immediately after sample collection, at the same time every day. Time of blood collection for each dog was randomly assigned by use of a computerized random-number generator.16
Blood sample collection—Blood samples (10 mL) were obtained from a jugular vein via careful veni-puncture with a 20-gauge needle attached to a 12-mL plastic syringe. On day 0, samples were immediately transferred into two 1.8-mL plastic tubes containing 3.2% sodium citratek (final blood-to-citrate ratio, 9:1), 1 plastic tube containing EDTA,l and 1 serum tube,m in that order. On days 2, 4, and 6, blood was transferred to 1 tube containing citrate and 1 tube containing EDTA as described; the tubes were then inverted gently 3 to 4 times after being filled to ensure anticoagulation.
Blood sample analysis—Samples for CBC, serum biochemical analysis, and coagulation profile were submitted to a laboratory,n and assays were performed with automated hematologic,o biochemical,p and coagulationq analyzers, respectively.
Citrated blood samples for TEG were stored for 30 minutes at room temperature (approx 21°C) until analysis.17 Samples were analyzed by use of a TEG analyzer.r All analyses were performed by 1 operator (SKF). For analysis, 20 μL of 0.2 mmol/L calcium chloride solutions was added to a TEG cup,t which was prewarmed to 37°C. A 340-μL aliquot of citrated blood was subsequently added to the cup, the pin was inserted, and TEG analysis softwareu was started. Testing was performed until the following variables had been determined: reaction time (R), clotting time (K), α-angle, MA, global clot strength (G), CI, and CL60. The variables G, CI, and CL60 are all mathematical calculations derived by the TEG software.r The G value is a mathematical transformation of MA, which can be interpreted as an actual measure of clot strength or shear elastic modulus strength, and was calculated by use of the formula G = (5,000 × MA/[100 − MA])/1,000.r The CI is an overall measure of coagulation status, calculated by use of the formula CI = −0.245(R) + 0.0184(K) + 0.1655(MA) − 0.0241 (α-angle) − 5.0220.18,19 The CL60 represents the percentage of MA that remains 60 minutes after MA is reached.u Reference intervals for platelet counts were obtained from the commercial laboratory that performed the countsn; reference intervals for TEG values were obtained from another study.20
Statistical analysis—For the preliminary experiment, statistical analysis was not performed because of the small sample size (n = 2). The MA for those 2 dogs was assessed to have returned to baseline values only if it was less than or equal to the day 0 value.
For the main study, analyses were performed by use of dedicated statistical software.v Normality was assessed by use of a Shapiro-Wilk test. Data expressed as percentages (CL60) were logit transformed for analysis. Parametric data are reported as mean ± SD, and logit-transformed data are expressed as median with 95% confidence interval. Data were analyzed by use of ANOVA for repeated measures, with the effects of treatment, time, treatment by time, and age entered into the model. Correlations between platelet count and the TEG variables MA and α-angle across and within all groups were performed by use of a Spearman correlation test. For all analyses, values of P < 0.05 were considered significant.
Results
Preliminary experiment—Both dogs completed the study, and no clinical signs of adverse effects were detected. Baseline values for both dogs were within the respective reference intervals for MA (57.7 mm in dog 1 and 54.3 mm in dog 2; reference interval, 44.4 to 61.9 mm) and G (6.83 kdyn/cm2 in dog 1 and 5.94 kdyn/cm2 in dog 2; reference interval, 3.7 to 7.6 kdyn/cm2). Although statistical comparisons were not made, values for MA and G were increased, compared with their respective baseline values, in both dogs in response to prednisone administration and remained greater than baseline for 4 weeks after prednisone treatment was discontinued. The MA exceeded the reference interval on days 4, 6, 13, and 27 in dog 1, with values ranging from 63.3 to 65.3 mm. In dog 2, MA exceeded the reference interval on days 6 and 27 (62.4 and 64.3 mm, respectively). Values for G exceeded the reference interval in dog 1 on days 2, 4, 6, 13, 20, and 27 (range of values above the reference interval, 7.90 to 9.32 kdyn/cm2) and in dog 2 on days 2, 6, 13, and 27 (7.92 to 9.00 kdyn/cm2). The remaining TEG variables evaluated (R, K, α-angle, CI, and CL60) varied relative to baseline values but did not exceed the respective reference intervals, except for a mild increase in CI values (reference interval, 0.5 to 2.9) in dog 1 on days 4, 6, and 13 (3.2, 3.0, and 3.2, respectively) and in dog 2 on day 27 (3.0). Values for CL60 were not obtained in dog 1 on days 0, 2, and 13 and in dog 2 on day 2.
Main study—All dogs completed the study without clinical signs of adverse effects, except for 1 dog in the prednisone group that developed diarrhea on day 5. Administration of prednisone was discontinued, and day 6 values for this dog were excluded from analysis.
All TEG variables except CL60 were normally distributed, and mean baseline values for TEG variables were not different among groups. Age did not affect TEG values. Overall treatment effects were evaluated by use of combined data for all time points within each treatment group. No differences were detected in mean (or median for CL60) values for any TEG variables between the ASA and control groups. In the prednisone and drug combination groups, mean values for K, α-angle, and CI were not significantly different from control group values, but mean values for MA and G and median CL were significantly increased (Table 1). Mean G in the drug combination group was greater than the upper limit of the reference interval.
Thromboelastography values* calculated by use of data measured in blood samples at all study time points (days 0 [baseline], 2, 4, and 6) in 16 healthy mixed-breed hounds grouped according to treatment (4 dogs/group).
| Group | |||||
|---|---|---|---|---|---|
| Variable | Reference interval† | Prednisone | ASA | Drug combination | Control |
| R(min) | 2.0 to 11.7 | 4.3 ± 1.4 | 4.1 ± 0.88 | 4.8 ± 1.4 | 3.9 ± 0.94 |
| K(min) | 1.1 to 5.8 | 2.2 ± 0.60 | 2.2 ± 0.45 | 2.0 ± 0.34 | 2.4 ± 0.63 |
| α-Angle(o) | 37.0 to 78.1 | 61.1 ± 6.0 | 60.2 ± 4.9 | 61.1 ± 5.6 | 59.7 ± 6.9 |
| MA (mm) | 44.4 to 61.9 | 58.8 ± 5.3‡ | 54.0 ± 5.9 | 60.1 ± 2.0§ | 53.0 ± 3.7 |
| G (kdyn/cm2) | 3.7 to 7.6 | 7.3 ± 1.4‡ | 6.0 ± 1.5 | 7.7 ± 1.4∥ | 5.7 ± 0.93 |
| CI | −0.5 to 2.9 | 2.2 ± 0.70 | 1.5 ± 0.98 | 2.3 ± 0.70 | 1.4 ± 0.55 |
| CL60 (%) | 78.2 to 100 | 96.3 (94.0–98.7)¶ | 89.0 (82.6–95.7) | 96.7 (94.6–98.8)§ | 89.7 (78.6–95.2) |
| Platelet count (× 109/L) | 117 to 418 | 281 ± 97 | 281 ± 31 | 262 ± 26 | 282 ± 55 |
Dogs received prednisone (median dose, 2.07 mg/kg), ASA (median dose, 0.51 mg/kg), or both drugs, PO, every 24 hours from days 0 through 6; the control group received no treatment. Baseline values were determined in blood samples (obtained immediately prior to administration of drugs in treated dogs) on day 0. Samples obtained on subsequent days were collected immediately prior to drug administration in dogs that received treatment. One dog in the prednisone group was censored because of potential adverse effects (diarrhea) on day 5; day 6 values did not include data from that dog.
Data for CL60 were logit transformed for analysis and are reported as median (95% confidence interval); all other data are reported as mean ± SD.
Reference intervals for platelet count were obtained from the testing laboratory,n and intervals for all other TEG values were described elsewhere.20
Within a variable, symbols indicate significant differences, compared with the control group value P = 0.03.
Within a variable, symbols indicate significant differences, compared with the control group value P = 0.01.
Within a variable, symbols indicate significant differences, compared with the control group value P = 0.007.
Within a variable, symbols indicate significant differences, compared with the control group value P = 0.02.
G = Global clot strength. K = Clotting time. R = Reaction time.
Data for all study dogs were combined for analysis of overall time effect by use of ANOVA (Table 2). Significant effects of time were detected for multiple variables; mean R values for these combined groups on days 2 and 6 were greater than day 0 baseline values. Mean MA was increased on days 2, 4, and 6, and mean G was increased on days 4 and 6, compared with the respective baseline values.
Thromboelastography values* and platelet counts on treatment days calculated by use of combined data for the 16 dogs in Table 1.
| Day of treatment | |||||
|---|---|---|---|---|---|
| Variable | Reference interval† | 0 (Baseline) | 2 | 4 | 6 |
| R(min) | 2.0 to 11.7 | 3.6 ± 1.0 | 4.7 ± 1.3‡ | 4.2 ± 1.0 | 4.5 ± 1.2§ |
| K(min) | 1.1 to 5.8 | 2.3 ± 0.64 | 2.4 ± 0.47 | 2.1 ± 0.50 | 2.2 ± 0.45 |
| α-Angle (o) | 37.0 to 78.1 | 60.6 ± 7.1 | 58.1 ± 5.4 | 61.7 ± 5.1 | 61.8 ± 5.1 |
| MA (mm) | 44.4 to 61.9 | 53.9 ± 6.1 | 56.2 ±4.6∥ | 58.0 ± 5.9¶ | 57.0 ± 5.6‡ |
| G (kdyn/cm2) | 3.7 to 7.6 | 6.0 ± 1.5 | 6.6 ± 1.3 | 7.1 ± 1.7¶ | 6.8 ± 1.6# |
| CI | −0.5 to 2.9 | 1.6 ± 0.87 | 1.8 ± 0.7 | 2.1 ± 0.86 | 1.9 ± 0.82 |
| CL60 (%) | 78.2 to 100 | 89.7 (88.3–95.1) | 94.5 (91.4–96.5) | 93.7 (90.1–96.0) | 96.7 (95.9–98.4)** |
| Platelet count (× 109/L) | 117 to 418 | 261 ± 58 | 263 ± 53 | 286 ± 49†† | 290 ± 62†† |
One dog in the prednisone group was censored because of potential adverse effects (diarrhea) on day 5; thus, n = 15 on day 6.
Within a variable, symbols indicate significant differences from the baseline value for all dogs P = 0.003
Within a variable, symbols indicate significant differences from the baseline value for all dogs P = 0.02.
Within a variable, symbols indicate significant differences from the baseline value for all dogs P = 0.04.
Within a variable, symbols indicate significant differences from the baseline value for all dogs P = 0.002.
Within a variable, symbols indicate significant differences from the baseline value for all dogs P = 0.03
Within a variable, symbols indicate significant differences from the baseline value for all dogs P = 0.001.
Within a variable, symbols indicate significant differences from the baseline value for all dogs P < 0.001.
See Table 1 for remainder of key.
Thromboelastography variables were evaluated to assess treatment-by-time effects (Table 3). Overall treatment-by-time effect was not significant for G (P = 0.07) or MA (P = 0.17). Although mean G exceeded the reference interval in the prednisone and drug combination groups and mean MA exceeded the reference interval in the drug combination group on days 4 and 6 of treatment, the increases were not significantly different from baseline values. Overall treatment-by-time effect was not significant (P = 0.18) for CI. There was a significant (P = 0.001) overall treatment-by-time effect for CL60. Decreased median clot lysis, as indicated by increased CL60 values, was detected on day 4 in the drug combination group and on day 6 in the prednisone and drug combination groups, compared with the group baseline and values in control dogs on the same day. There was no significant overall treatment-by-time effect on R (P = 0.24).
Thromboelastography values* and platelet counts over time for the 16 dogs in Table 1 grouped according to treatment (4 dogs/group).
| Day of treatment | |||||
|---|---|---|---|---|---|
| Variable | Group | 0 (Baseline) | 2 | 4 | 6 |
| MA (mm) | Prednisone | 53.7 ± 6.4 | 58.2 ± 3.7 | 61.5 ± 2.8 | 61.0 ± 1.1 |
| ASA | 52.4 ± 7.5 | 56.5 ± 7.0 | 54.3 ± 6.3 | 52.7 ± 3.4 | |
| Drug combination | 56.4 ± 6.6 | 57.3 ± 2.8 | 63.1 ± 3.9 | 62.6 ± 2.6 | |
| Control | 53.2 ± 5.9 | 52.6 ± 3.9 | 53.3 ± 3.7 | 52.8 ± 1.8 | |
| G (kdyn/cm2) | Prednisone | 6.0 ± 1.5 | 7.0 ± 1.0 | 8.0 ± 0.9 | 8.0 ± 1.7 |
| ASA | 5.7 ± 1.9 | 6.7 ± 2.1 | 6.1 ± 1.4 | 5.6 ± 0.8 | |
| Drug combination | 6.6 ± 1.6 | 6.8 ± 0.8 | 8.6 ± 1.4 | 8.4 ± 0.9 | |
| Control | 5.8 ± 1.5 | 5.7 ± 0.9 | 5.8 ± 0.9 | 5.6 ± 0.4 | |
| CI | Prednisone | 1.8 ± 0.7 | 2.0 ± 0.6 | 2.6 ± 0.5 | 2.3 ± 0.9 |
| ASA | 1.1 ± 1.5 | 1.9 ± 1.2 | 1.7 ± 1.1 | 1.3 ± 0.5 | |
| Drug combination | 1.9 ± 0.8 | 1.8 ± 0.4 | 2.7 ± 0.6 | 2.7 ± 0.4 | |
| Control | 1.5 ± 0.9 | 1.4 ± 0.5 | 1.4 ± 0.5 | 1.3 ± 0.4 | |
| CL60 (%) | Prednisone | 93.3 (81.4–97.1) | 95.2 (90.8–98.5) | 95.6 (86.4–98.0) | 99.8 (99.0–99.9)†‡ |
| ASA | 90.2 (83.7–97.1) | 90.3 (82.4–96.8) | 88.1 (75.9–95.3) | 89.6 (78.0–95.8) | |
| Drug combination | 93.4 (86.9–97.7) | 95.5 (91.3–98.6) | 96.6 (95.5–99.3)§∥ | 96.9 (97.1–99.5)¶# | |
| Control | 89.2 (75.0–95.5) | 91.4 (79.0–96.8) | 89.1 (69.9–94.6) | 90.7 (78.6–96.3) | |
| R (min) | Prednisone | 2.7 ± 0.5 | 5.1 ± 0.8 | 4.4 ± 1.0 | 5.1 ± 0.1 |
| ASA | 4.4 ± 0.7 | 4.4 ± 0.7 | 3.6 ± 0.6 | 4.2 ± 1.4 | |
| Drug combination | 4.1 ± 1.3 | 5.6 ± 1.0 | 4.8 ± 1.5 | 4.8 ± 1.9 | |
| Control | 3.4 ± 0.6 | 3.8 ± 1.4 | 4.1 ± 0.7 | 4.1 ± 1.1 | |
| Platelet count (cells × 109/L) | Prednisone | 245 ± 108 | 269 ± 100 | 330 ± 56** | 294 ± 137†† |
| ASA | 260 ± 21 | 261 ± 22 | 295 ± 28‡‡ | 311 ± 24‡ | |
| Drug combination | 248 ± 32 | 260 ± 14 | 273 ± 37§§ | 269 ± 18 | |
| Control | 287 ± 47 | 262 ± 61 | 284 ± 72 | 298 ± 57 | |
One dog in the prednisone group was censored because of potential adverse effects (diarrhea) on day 5; thus, n = 3 for that group on day 6.
Within a variable, symbols indicate significant differences, compared with the control group value on the same day P < 0.001.
Within a variable, symbols indicate significant differences, compared with the control group value on the same day P = 0.004.
Within a variable, symbols indicate significant differences, compared with the control group value on the same day P = 0.003.
Within a variable, symbols indicate significant differences from the baseline value for the same group P < 0.001.
Within a variable, symbols indicate significant differences from the baseline value for the same group P = 0.03;.
Within a variable, symbols indicate significant differences from the baseline value for the same group P = 0.007.
Within a variable, symbols indicate significant differences from the baseline value for the same group P = 0.001.
Within a variable, symbols indicate significant differences from the baseline value for the same group P = 0.01.
Within a variable, symbols indicate significant differences from the baseline value for the same group P = 0.002.
Within a variable, symbols indicate significant differences from the baseline value for the same group P = 0.02.
See Table 1 for remainder of key.
Significant time (P < 0.001) and overall treatment-by-time (P = 0.03) effects were determined for the platelet count. Mean platelet counts were increased in combined data for all dogs on days 4 and 6 of treatment, compared with the baseline value on day 0 (Table 2). Mean platelet counts were increased in all treatment groups except the control group on day 4 and were increased in the prednisone and ASA groups on day 6, compared with the respective baseline values (Table 3). Despite these increases, mean platelet counts remained within the reference interval of 117 × 109/L to 418 × 109/L.n
No significant correlations were identified between platelet count and MA in the prednisone (r = −0.07; P = 0.8), ASA (r = 0.14; P = 0.6), drug combination (r = 0.36; P = 0.17), or control (r = 0.29; P = 0.3) groups. These correlations were underpowered; power was 2.91%, 0.08%, 28.22%, and 19.24% in prednisone, ASA, drug combination, and control groups, respectively. A fair correlation was detected between mean platelet count and mean MA when data for all groups at all time points were combined (r = 0.30; P = 0.02). In addition, platelet count was moderately correlated with α-angle when data for all groups at all time points were combined (r = 0.51; P < 0.001).
Discussion
In the study reported here, we investigated the independent and combined effects of prednisone and ASA on TEG variables in healthy dogs. Treatment with prednisone, with and without ASA, resulted in increased clot strength and decreased clot lysis. The increase in clot strength was indicated by increased mean MA and G values in the prednisone and drug combination groups as an overall treatment effect. However, the overall treatment-by-time effects for these variables were not significant. The lack of significance of the treatment-by-time effect was likely attributable to insufficient group size. Significant increases in mean MA and G were detected on days 4 and 6 when data for all groups were combined, and the authors believe that this effect was predominantly caused by the contribution of the prednisone and drug combination group values to the overall means. The value of G is mathematically calculated from MA, with G = (5,000 × MA/[100 − MA])/1,000; therefore, G is more sensitive to small changes in clot strength than is MA. Measurement of G did not add substantially to the analysis in the present study but was included because it was used to classify patients in other veterinary studies21–24 as having a hypercoagulable state on the basis of a value > 7.2 kdyn/cm2 in tissue-factor–activated TEG analysis. It is not known what degree of change in either MA or G corresponds to an increased risk of TED in dogs, nor whether the increased sensitivity for small changes provided by G is clinically relevant.
The mechanisms of hypercoagulability in dogs treated with corticosteroids are not fully understood. Changes in concentrations of coagulation factors and natural anticoagulants have been reported. Dogs with naturally occurring hyperadrenocorticism have decreased plasma antithrombin activity; increased plasma activities of factors II, V, VII, IX, X, and XII; and increased plasma fibrinogen concentrations.25 It is not known whether exogenously administered corticosteroids induce the same changes in dogs. However, in clinically normal humans treated with dexamethasone for 5 days, increased plasma activities of factors VII, VIII, and XI and increased plasma fibrinogen concentrations were observed in 1 study.26 Similar effects were observed in humans that received methylprednisolone treatment in another study,27 except that a decrease in plasma fibrinogen concentrations was detected after treatment.
Corticosteroids and hyperadrenocorticism are commonly believed to cause thrombocytosis; however, reactive thrombocytosis is not believed to result in TED in the absence of other risk factors, such as neoplasia.28,29 Thrombocytosis may cause MA to increase; however, in the present study, it is unlikely that the increases in mean MA and G values detected as a treatment effect between groups were related to corresponding increases in platelet count because platelet count was not correlated with MA for any of the 4 groups. Because the correlation analyses were underpowered, an effect may have been revealed if additional dogs were included in the study. When all groups were combined for correlation analysis, platelet count was correlated with MA and α-angle. Maximum amplitude is commonly cited as the variable most affected by platelet count and function, and it seems intuitive that MA and platelet count should be correlated.8,30–32 The α-angle represents the speed of fibrin cross-linking, and as this process occurs on the platelet procoagulant surface, it is likely that a change in platelet number may affect speed of clot formation.8 Mean platelet count was increased relative to baseline values in the prednisone, ASA, and drug combination groups on day 4 and in the prednisone and ASA groups on day 6. These changes are not likely to be attributable to corticosteroid treatment alone, as the greatest increase in platelet count was in the ASA group on day 6. One possible explanation33 for this change in platelet count over time is epinephrine-induced splenic contraction and release of stored platelets into circulation in dogs anticipating venipuncture on later days of the study. Alternatively, the described increase in mean platelet count may have been the result of analytic variation, as microaggregates of platelets have the potential to artificially decrease platelet counts.33
The increase in median CL60 in dogs treated with prednisone in the present study was attributed to a decrease in fibrinolysis, relative to baseline values and those of control dogs. Studies evaluating the effects of corticosteroids on fibrinolysis have conflicting results. Results of a study34 in humans with various inflammatory diseases revealed increased fibrinolysis from baseline values, demonstrated by decreased clot lysis times, in patients treated with corticosteroids. In another study,35 no effect was found on markers of fibrinolysis in healthy volunteers treated with high doses of corticosteroids, and yet another study26 revealed a mild decrease in plasma concentrations of plasminogen activator inhibitor-1, compared with those of individuals that received a placebo control. However, in the latter 2 studies, patients received treatment for only 5 days, which may not be long enough to demonstrate suppression of fibrinolysis. Investigators of studies36–38 in which long-term corticosteroid treatment was evaluated following kidney and heart transplantation reported significantly upregulated plasminogen activator inhibitor-1 plasma antigen concentrations and activity over time with decreased fibrinolytic activity. Such patients are considered to be at high risk of cardiac and cerebral injury caused by TED, and impaired fibrinolysis is thought to play a role.36 The effect of corticosteroids on fibrinolysis has not been described in dogs, nor have TEG clot lysis variables been compared with other means of assessing fibrinolysis, such as euglobulin lysis time, or circulating concentrations of tissue plasminogen activator or plasminogen activator inhibitor-1. Rotational thromboelastomometry, a technology closely related to TEG, was demonstrated to be a sensitive indicator of fibrinolysis in humans and was correlated to more traditional means of evaluating fibrinolysis.39
Although prednisone administration in the study reported here resulted in the described increased mean MA and G and increased median CL60 in the prednisone and drug combination groups, the changes were mild. It is commonly stated that an increase in MA or G and increased CL60 indicate hypercoagulability, but it is unknown what values represent a clinically relevant increase in risk for TED. Dogs in the present study did not have any clinical signs of TED, but they were healthy prior to receiving treatment and were only treated for 7 days. In studies9,c,d of dogs with IMHA, which are at increased risk for TED, the reported increase in MA above laboratory reference intervals9,d and values in control dogsc was greater than that detected for healthy dogs treated with prednisone in the present study. It is not known whether treatment with prednisone could further increase the risk for TED in dogs with IMHA or whether any effect of the drug would be clinically irrelevant in the presence of preexisting hypercoagulability.
Treatment with ASA did not have any effect on TEG variables in the ASA group, nor did it ameliorate the effects of prednisone in the drug combination group. Decreased platelet function should result in decreased MA and G.8 In a study40 in humans, an increased K value was found to be a sensitive indicator of platelet dysfunction. The lack of effect of ASA on TEG in the present study is not surprising. Treatment with ASA at a much higher dosage (5 mg/kg, q 12 h) in a study41 in dogs with osteoarthritis did not alter TEG values, although it did result in decreased platelet aggregation, compared with baseline values, as detected by means of optical aggregometry. Another report42 indicated that ASA administered at a dosage of 10 mg/kg every 12 hours did not affect closure times assessed via optical platelet aggregometry or by use of a platelet function analyzer in healthy dogs. A low dosage of ASA (0.5 mg/kg/d) has been variably shown to alter platelet function. In healthy dogs, results of optical aggregometry were not affected by this dosage of ASA in 1 report,a but in another, whole blood impedance aggregometry results were affected by the same dosage.b The discrepancies between study results may be a result of different platelet agonists and sample types or methods (ie, optical aggregometry vs impedance aggregometry) and possibly different individual responses to ASA treatment. Breed differences in platelet aggregation following ASA treatment were revealed by use of platelet function analyzer closure times and whole blood impedance aggregometry.43 It should be noted, however, that TEG as used in the present study is less sensitive than is platelet aggregometry for the detection of mild platelet dysfunction. Fibrin contributes to clot strength and therefore to MA and G, such that fibrin may mask the effects of platelet dysfunction.44 To more accurately assess platelet dysfunction with TEG, there is a modified TEG method that more accurately reflects platelet function and is well correlated to optical aggregometry in humans44,45; this assay has been shown to detect platelet inhibition in response to clopidogrel treatment in dogs, and the technique should be validated for use in monitoring ASA treatment in dogs.46
The prolongation (increased values) in mean R in dogs of the present study over time is difficult to explain. The R is a reflection of coagulation factor function, and hypercoagulability is typically indicated by a shortened R.8,30 As such, we would have expected dogs treated with prednisone to have a shortened R after treatment. In fact, the opposite effect was observed and the mean R on days 2 and 6 was significantly longer than that measured at baseline in the combined group data. There are several possible explanations for this variation in R over time. Preanalytic error, such as inadequate mixing of the blood sample with citrate, may have led to incomplete inactivation of coagulation by citrate and in vitro contact-activated thrombin generation.47 This could potentially lead to a shorter R on day 0 caused by preexisting thrombin in the sample and more rapid platelet activation in vitro. Analytic error could have also played a role if the TEG was not started immediately after addition of citrate to the blood, which would have shortened R on day 0, or incomplete recalcification of the blood during analysis could have prolonged the R on days 2 and 6. Because there was no evidence for preanalytic or analytic error, the change in mean R could reflect biological variation; none of these values were outside of the reference interval for R in recalcified whole blood.
The major limitation of the present study was small group size. The preliminary experiment, in which prednisone (2 mg/kg, PO, q 24 h, for 7 days) was administered to 2 clinically normal Beagles, revealed that apparent increases in MA and G, compared with baseline values, persisted for at least 4 weeks after discontinuation of treatment. This corroborated similar findings in another studyw in which investigators evaluated prednisone treatment in healthy Beagles; in that study, TEG changes persisted for up to 6 weeks after treatment with prednisone at 1 and 4 mg/kg, PO, every 24 hours for 2 weeks. As such, a Latin square design, which would have increased power with the small number of dogs in the present study, was not feasible because the washout period that would have been necessary was undetermined. The group size of 4 dogs was sufficient to reveal increases in mean MA and G values and an increase in median CL60 in response to prednisone treatment but may have been responsible for the lack of detectable changes in other TEG variables after prednisone administration and lack of detectable effects of ASA treatment. The possibility that effects of prednisone on TEG variables in the present study were the result of type I error attributable to small group size was considered to be unlikely; the findings reported here were consistent with those of the previously described studies of corticosteroid effects on hemostasis in humans and dogs. Another limitation of the study reported here was the short duration of treatment. Most dogs with IMHA or other immune-mediated diseases are treated for weeks to months, and it is unknown whether the hypercoagulability detected in dogs of the present study would become more severe with time. Also, if the changes determined by use of TEG in the present study did become more pronounced with a longer treatment period, detection of these changes could possibly have improved with this small number of dogs.
In the present study, the effects of prednisone and TEG were evaluated in healthy male dogs only. An improved understanding of the effects of these medications in healthy animals may aid in interpretation of TEG variables in patients treated with these drugs, including dogs with IMHA.
ABBREVIATIONS
| ASA | Acetylsalicylic acid |
| CI | Coagulation index |
| CL60 | Percentage of clot lysis at 60 minutes |
| IMHA | Immune-mediated hemolytic anemia |
| MA | Maximum amplitude |
| TED | Thromboembolic disease |
| TEG | Thromboelastography |
Shearer L, Kruth SJ, Wood D. Effects of aspirin and clopidogrel on platelet function in healthy dogs (abstr). J Vet Intern Med 2009;23:745.
Sharpe KS, Center SA, Randolph JF, et al. Platelet impedance aggregometry in clinically healthy dogs with and without ultralow-dose aspirin (abstr). J Vet Intern Med 2009;23:693.
Goggs R, Wiinberg B, Kjelgaard-Hansen M, et al. Serial analysis of coagulation parameters in dogs with immune-mediated hemolytic anemia (IMHA) using thromboelastography (TEG) (abstr). J Vet Intern Med 2010;24:680.
Flint S, Abrams-Ogg A, Kruth S, et al. Thromboelastography in dogs with immune-mediated haemolytic anemia treated with prednisone, azathioprine and low-dose aspirin (abstr). J Vet Intern Med 2010;24:681.
Vilar P, Hansell J, Westendorf N, et al. Effects of hematocrit on thromboelastography tracings in dogs (abstr). J Vet Intern Med 2008;22:774.
Prednisone 5-mg tablets, Novopharm Ltd, Toronto, ON, Canada.
Prednisone 50-mg tablets, Novopharm Ltd, Toronto, ON, Canada.
Asaphen, Pharmascience Inc, Montreal, QC, Canada.
Ontario Veterinary College Teaching Hospital Pharmacy, University of Guelph, Guelph, ON, Canada.
Gel Capsules, Tub Enterprises, Almonte, ON, Canada.
BD Vacutainer Citrate Tube, BD, Franklin Lakes, NJ.
BD Vacutainer EDTA Tube, BD, Franklin Lakes, NJ.
BD Vacutainer Serum Tube, BD, Franklin Lakes, NJ.
Clinical Pathology lab section, Animal Health Laboratory, Laboratory Services Division, University of Guelph, Guelph, ON, Canada.
Advia 120, Siemens Healthcare Diagnostics, Mississauga, ON, Canada.
Cobas C 501, Roche Diagnostics Canada, Laval, QC, Canada.
Amelung KC4, Trinity Biotech USA, Jamestown, NY.
TEG 5000 Thrombelastograph Hemostasis Analyser, Haemoscope Corp, Niles, Ill.
Calcium Chloride 0.2M, Haemoscope Corp, Niles, Ill.
Disposable Cups and Pins, Haemoscope Corp, Niles, Ill.
TEG Analytical Software, version 4.2.3, Haemoscope Corp, Niles, Ill.
SAS, version 9.1.3, SAS Institute Inc, Cary, NC.
Rose L, Bedard C, Dunn M. Effect of prednisone administration on thromboelastography parameters in healthy dogs (abstr). J Vet Intern Med 2008;22:738–739.
References
- 1.
Carr AP, Panciera DL, Kidd L. Prognostic factors for mortality and thromboembolism in canine immune-mediated hemolytic anemia: a retrospective study of 72 dogs. J Vet Intern Med 2002; 16: 504–509.
- 2.
Johnson LR, Lappin MR, Baker DC. Pulmonary thromboembolism in 29 dogs: 1985–1995. J Vet Intern Med 1999; 13: 338–345.
- 3.↑
Weinkle TK, Center SA, Randolph JF, et al. Evaluation of prognostic factors, survival rates, and treatment protocols for immune-mediated hemolytic anemia in dogs: 151 cases (1993–2002). J Am Vet Med Assoc 2005; 226: 1869–1880.
- 4.
Balch A, Mackin A. Canine immune-mediated hemolytic anemia: treatment and prognosis. Compend Contin Educ Pract Vet 2007; 29: 230–238.
- 5.
LaRue MJ, Murtaugh RJ. Pulmonary thromboembolism in dogs: 47 cases (1986–1987). J Am Vet Med Assoc 1990; 197: 1368–1372.
- 6.
Burns MG, Kelly AB, Hornof WJ, et al. Pulmonary artery thrombosis in three dogs with hyperadrenocorticism. J Am Vet Med Assoc 1981; 178: 388–393.
- 7.
Boswood A, Lamb CR, White RN. Aortic and iliac thrombosis in six dogs. J Small Anim Pract 2000; 41: 109–114.
- 8.↑
Donahue SM, Otto CM. Thromboelastography: a tool for measuring hypercoagulability, hypocoagulability, and fibrinolysis. J Vet Emerg Crit Care 2005; 15: 9–16.
- 9.↑
Sinnott VB, Otto CM. Use of thromboelastography in dogs with immune-mediated hemolytic anemia: 39 cases (2000–2008). J Vet Emerg Crit Care 2009; 19: 484–488.
- 10.
Kheirabadi BS, Crissey JM, Deguzman R, et al. Effects of synthetic versus natural colloid resuscitation on inducing dilutional coagulopathy and increasing hemorrhage in rabbits. J Trauma 2008; 64: 1218–1228.
- 11.
Spiezia L, Radu C, Marchioro P, et al. Peculiar whole blood rotation thromboelastometry (rotem) profile in 40 sideropenic anaemia patients. Thromb Haemost 2008; 100: 1106–1110.
- 12.↑
Hobson AR, Agarwala RA, Swallow RA, et al. Thrombelastography: current clinical applications and its potential role in interventional cardiology. Platelets 2006; 17: 509–518.
- 13.↑
Pittman JR, Koenig A, Brainard BM. The effect of unfractionated heparin on thrombelastographic analysis in healthy dogs. J Vet Emerg Crit Care 2010; 20: 216–223.
- 14.
Olfert ED, Cross BM, McWilliam AA, eds. Guide to the care and use of experimental animals. Vol 1. 2nd ed. Ottawa: Canadian Council on Animal Care, 1993.
- 15.
Olfert ED, Cross BM, McWilliam AA, eds. Guide to the care and use of experimental animals. Vol 2. Ottawa: Canadian Council on Animal Care, 1984.
- 16.↑
Fishman GS, Moore LR. A statistical evaluation of multiplicative congruential random number generators with modulus 231 – 1. J Am Stat Assoc 1982; 77: 129–136.
- 17.↑
Wiinberg B, Jensen AL, Rojkjaer R, et al. Validation of human recombinant tissue factor-activated thromboelastography on citrated whole blood from clinically healthy dogs. Vet Clin Pathol 2005; 34: 389–393.
- 18.
Caprini JA, Zuckerman L, Cohen E, et al. The identification of accelerated coagulability. Thromb Res 1976; 9: 167–180.
- 19.
Cohen E, Caprini J, Zuckerman L, et al. Evaluation of three methods used to identify accelerated coagulability. Thromb Res 1977; 10: 587–604.
- 20.↑
Flint SK, Wood RD, Abrams-Ogg ACG, et al. Comparison of citrated native and kaolin-activated thromboelastography parameters in healthy dogs during development of a reference interval. Vet Clin Pathol 2011; 40:in press.
- 21.
Kristensen AT, Wiinberg B, Jessen LR, et al. Evaluation of human recombinant tissue factor-activated thromboelastography in 49 dogs with neoplasia. J Vet Intern Med 2008; 22: 140–147.
- 22.
Vilar P, Couto CG, Westendorf N, et al. Thromboelastographic tracings in retired racing Greyhounds and in non-Greyhound dogs. J Vet Intern Med 2008; 22: 374–379.
- 23.
Wiinberg B, Jensen AL, Rozanski E, et al. Tissue factor activated thromboelastography correlates to clinical signs of bleeding in dogs. Vet J 2009; 179: 121–129.
- 24.
Wiinberg B, Jensen AL, Johansson PI, et al. Thromboelastographic evaluation of hemostatic function in dogs with disseminated intravascular coagulation. J Vet Intern Med 2008; 22: 357–365.
- 25.↑
Jacoby RC, Owings JT, Ortega T, et al. Biochemical basis for the hypercoagulable state seen in Cushing syndrome. Arch Surg 2001; 136: 1003–1006.
- 26.↑
Brotman DJ, Girod JP, Posch A, et al. Effects of short-term glucocorticoids on hemostatic factors in healthy volunteers. Thromb Res 2006; 118: 247–252.
- 27.↑
Frank RD, Altenwerth B, Brandenburg VM, et al. Effect of intravenous high-dose methylprednisolone on coagulation and fibrinolysis markers. Thromb Haemost 2005; 94: 467–468.
- 28.
Brooks MB, Catalfamo JL. Immune-mediated thrombocytopenia, von Willebrand disease, and platelet disorders. In: Ettinger SJ, Feldman EC, eds. Textbook of veterinary internal medicine: diseases of the dog and the cat. 7th ed. St Louis: Saunders Elsevier, 2010;772–783.
- 29.
Vannucchi AM, Barbui T. Thrombocytosis and thrombosis. Hematol Am Soc Hematol Educ Program2007;363–370.
- 30.
Traverso CI, Caprini JA, Arcelus JI. The normal thromboelastogram and its interpretation. Semin Thromb Hemost 1995; 21 (suppl 4:)7–13.
- 31.↑
Bowbrick VA, Mikhailidis DP, Stansby G. Influence of platelet count and activity on thromboelastography parameters. Platelets 2003; 14: 219–224.
- 32.
Oshita K, Az-ma T, Osawa Y, et al. Quantitative measurement of thromboelastography as a function of platelet count. Anesth Analg 1999; 89: 296–299.
- 33.↑
Stockham SL, Scott MA. Platelets. In: Stockham SL, Scott MA, eds. Fundamentals of veterinary clinical pathology. 2nd ed. Ames, Iowa: Blackwell, 2008;223–257.
- 34.↑
van Giezen JJ, Jansen JW. Inhibition of fibrinolytic activity in-vivo by dexamethasone is counterbalanced by an inhibition of platelet aggregation. Thromb Haemost 1992; 68: 69–73.
- 35.↑
Chakrabarti R, Fearnley GR, Hocking ED. Effect of corticosteroid therapy on fibrinolysis in patients with inflammatory and non-inflammatory conditions. Br Med J 1964; 29: 534–537.
- 36.↑
Sartori MT, Rigotti P, Marchini F, et al. Plasma fibrinolytic capacity in renal transplant recipients: effect of steroid-free immuno-suppression therapy. Transplantation 2003; 75: 994–998.
- 37.
Patrassi GM, Sartori MT, Livi U, et al. Impairment of fibrinolytic potential in long-term steroid treatment after heart transplantation. Transplantation 1997; 64: 1610–1614.
- 38.
Patrassi GM, Sartori MT, Rigotti P, et al. Reduced fibrinolytic potential one year after kidney transplantation relationship to long-term steroid treatment. Transplantation 1995; 59: 1416–1420.
- 39.↑
Levrat A, Gros A, Rugeri L, et al. Evaluation of rotation thromboelastography for the diagnosis of hyperfibrinolysis in trauma patients. Br J Anaesth 2008; 100: 792–797.
- 40.↑
Bowbrick VA, Mikhailidis DP, Stansby G. Value of thromboelastography in the assessment of platelet function. Clin Appl Thromb Hemost 2003; 9: 137–142.
- 41.↑
Brainard BM, Meredith CP, Callan MB, et al. Changes in platelet function, hemostasis, and prostaglandin expression after treatment with nonsteroidal anti-inflammatory drugs with various cyclooxygenase selectivities in dogs. Am J Vet Res 2007; 68: 251–257.
- 42.↑
Blois SL, Allen DG, Wood RD, et al. Effects of aspirin, carprofen, deracoxib, and meloxicam on platelet function and systemic prostaglandin concentrations in healthy dogs. Am J Vet Res 2010; 71: 349–358.
- 43.↑
Nielsen LA, Zois NE, Pedersen HD, et al. Platelet function in dogs: breed differences and effect of acetylsalicylic acid administration. Vet Clin Pathol 2007; 36: 267–273.
- 44.↑
Craft RM, Chavez JJ, Bresee SJ, et al. A novel modification of the thrombelastograph assay, isolating platelet function, correlates with optical platelet aggregation. J Lab Clin Med 2004; 143: 301–309.
- 45.
Swallow RA, Agarwala RA, Dawkins KD, et al. Thromboelastography: potential bedside tool to assess the effects of antiplatelet therapy? Platelets 2006; 17: 385–392.
- 46.↑
Brainard BM, Kleine SA, Papich MG, et al. Pharmacodynamic and pharmacokinetic evaluation of clopidogrel and the carboxylic acid metabolite SR 26334 in healthy dogs. Am J Vet Res 2010; 71: 822–830.
- 47.↑
Camenzind V, Bombeli T, Seifert B, et al. Citrate storage affects thrombelastograph analysis. Anesthesiology 2000; 92: 1242–1249.
