Hyperadrenocorticism (HAC) has been associated with thrombotic disease in dogs.
Hyperadrenocorticism (HAC) has been associated with thrombotic disease in dogs.
The purpose of this study was to use thromboelastography (TEG) and measurement of thrombin generation (TG) to characterize the hypercoagulable state in dogs with HAC. We hypothesized that dogs with HAC would have a hypercoagulable profile on TEG tracings and an increase in thrombin generation as measured by endogenous thrombin potential (ETP).
Sixteen dogs with HAC. Dogs were compared with a population of normal dogs used to obtain reference intervals.
TEG tracings on citrated whole blood were obtained from 15 dogs, and TG measurements on frozen-thawed platelet-poor plasma (PPP) were obtained from 15 dogs.
For the TEG analysis, when results of individual dogs were compared with the reference interval, 12/15 dogs had at least 1 parameter associated with hypercoagulability. When the population of HAC dogs was compared with a population of healthy dogs, HAC dogs had decreases in R and K and increases in α and MA values. The ETP was increased when the HAC group was compared with a population of normal dogs. However, only 3/15 dogs had an ETP above reference interval, and 1/15 had a decreased lag time.
Of 16 dogs with HAC, 12/15 had evidence of hypercoagulability when evaluated by TEG, 4/15 when evaluated by TG, and 2 dogs had increases in ETP and MA.
calibrated automated thrombogram
Complete blood cell count
endogenous thrombin potential
global clot strength
time to peak
Increased blood cortisol concentration associated with Cushing's syndrome or chronic exogenous administration has been associated with arterial thromboembolism. Detection of hypercoagulability in dogs remains a challenge. Global models of hemostasis such as thromboelastography (TEG) and thrombin generation (TG) have been suggested to provide a better assessment of hypercoagulability. Thrombin plays a central role in hemostasis and thrombosis. The calibrated automated thrombogram (CAT) is an assay that allows measurement of endogenous thrombin potential (ETP) in clotting plasma. Our goal was to characterize TEG and TG measurements in dogs with HAC.
The study was a prospective clinical evaluation conducted on client-owned dogs presented to the internal medicine service of the CHUV (Centre Hospitalier Universitaire Vétérinaire) and newly diagnosed with HAC. Dogs were excluded if they had been treated with exogenous corticosteroids (iatrogenic HAC), treated for HAC, or had received nonsteroidal anti-inflammatory drugs.
The values used as a reference interval for TEG were obtained from a control group of 31 healthy adult dogs (12 Beagles and 19 dogs of various breeds aged 7 months–10.5 years). The values used as a reference interval for TG were obtained from a control group of 20 healthy adult Beagle dogs.
Samples were obtained for TEG analysis and TG at the time of diagnosis of each patient.
Blood was collected by clean jugular venipuncture using a 21G vacutainer needle. To clear possible tissue factor arising from jugular trauma, 2.7 mL of blood was collected in a dry tube,1 and then 4 citrated tubes2 were filled at a blood-to-citrate ratio of 9:1. Two tubes were gently inverted 10 times and then left at room temperature for 30 minutes before performing TEG assays. The remaining tubes were centrifuged at 3,300×g for 10 minutes at 7°C. PPP was collected, placed in plastic tubes (aliquots of 1 cc), and immediately frozen and stored at −80°C until TG. The longest the samples were stored was 31 months, with an average storage time of 23 months.
Thromboelastography studies were performed on whole blood 30 minutes after blood collection using a Thromboelastograph Hemostasis Analyzer 5000 (TEG).3 Coagulation was initiated using recombinant tissue factor4 diluted 1/100 in phosphate-buffered saline (PBS) solution containing 4% bovine serum albumin, pH 7.4. Briefly, TEG assays were performed by adding 10 μL of tissue factor and 20 μL of CaCl2 (0.2 M) mixed with 330 μL of citrated whole blood. Tracings were run at 37°C for 120 minutes in duplicate. Values for global clot strength (G), reaction time (R), clotting time (K), angle (α), maximum amplitude (MA), and lysis 30 (LY30) were recorded. Dogs were identified as having abnormal TEG tracings when 1 or more of the TEG results fell outside of the minimum and maximum values obtained for control dogs. Dogs were considered to have hypercoagulable tracings when there was increased MA, decreased K, decreased α, or a combination thereof.
All samples were run with the fluorogenic calibrated automated thrombogram (CAT) assay by Thrombinoscope using the technique described by Hemker. Each experiment required 2 sets of readings, 1 from wells in which thrombin generation (TG) took place and 1 from wells to which the thrombin calibrator (Cal) was added. A volume of 80 μL PPP was added to 20 μL of the buffer solution containing the trigger solution in the TG wells, or to 20 μL of the thrombin calibrator in the Cal wells. The trigger solution contained a mixture of phospholipids and human recombinant tissue factor (provided by Thrombinoscope5). Thrombin generation was initiated by adding 20 μL of a solution containing calcium (CaCl2) in a Hepes buffer (pH 7.35) and the fluorescent substrate to the TG and Cal wells. The final solution contained 5 ρM of tissue factor and 4 μM of phospholipids. Fluorescence was measured using an automated plate reader fluorometer,6 using specific fluorescence plates7 and dedicated software.5 All analyses were performed in triplicate as recommended by the manufacturer. Values for lag time in minutes, endogenous thrombin potential (ETP) in nM/min, peak in nM of thrombin, and time to peak (TTpeak) in minutes were recorded (Fig 1).
The mean of the duplicates for each individual dog was compared to established reference intervals for all parameters for TEG (R, K, α, MA, and LY30) and TG (ETP, lag time, peak, and TTpeak). Values outside of the reference interval were considered to be abnormal. The reference intervals are those used by our coagulation laboratory. Platelet counts were evaluated retrospectively using the data available at the time of diagnosis, but no statistical analyses were performed. Unpaired t-tests for unequal variance were used to compare mean values of the HAC dogs and mean values of healthy dogs used to establish a reference interval for both TEG and TG parameters. Normality of these 2 groups was tested by an Anderson-Darling analysis that did not indicate significant deviation from normal. Statistical significance was set at P < .05 for all values. Analyses were carried out using SAS v.9.2.8
Sixteen dogs were included in the study: 10 spayed females and 6 castrated males with a mean weight of 18.2 kg (range, 4.7–53.6 kg) and mean age of 10.5 years (range, 6–13.5 years). Of these dogs, 13 had pituitary HAC, 2 had adrenal HAC, and 1 had HAC of unknown origin. Of these 16 dogs, 15 had TEG analysis and 15 had TG analysis done before treatment. No dog showed any clinical sign of thromboembolism during the study period.
When the mean TEG values of individual dogs were compared with the reference interval used in our coagulation laboratory (R [minutes], 0.70–2.44; K [minutes], 0.85–5.15; Alpha [degrees], 41.73–81.2; MA [mm], 38.72–64.28; LY30 [%], 0.00–9.30), the following results were found: no dog had an R value outside of the reference interval, 10/15 dogs had a K value lower than the reference interval, 1/15 dogs had an alpha value above the reference interval, 11/15 dogs had MA above the reference interval, and all dogs had LY30 values within the reference interval. Therefore, 8/15 dogs had 2 parameters that indicated hypercoagulability and 4/15 had 1 parameter that indicated hypercoagulability (Fig 2).
When the mean TG values of individual dogs were compared with the reference interval established by our coagulation laboratory (ETP [nM/min], 216–465; lag time [min], 0.84–1.67; peak [nM], 79.93–178.73; time to peak [min], 2.2–4.22), the following results were found: 3/15 had an ETP above the reference interval, 2/15 had a lag time outside of the reference interval (1 above and 1 below reference interval), 4/15 had a peak above reference interval, and no dog had a time to peak outside of the reference interval. Overall, 5/15 dogs had at least one TG value outside of the reference interval, 4 of which were compatible with hypercoagulability (Fig 3).
When platelet counts were evaluated, 11/14 dogs had platelet counts above the reference range (143–400 × 109/L). No patient was considered to have a platelet count below reference range.
Mean values for TEG and standard deviations are presented in Table 1. When the HAC mean values were compared to the mean of the reference interval, it was found that in the HAC patients, R was significantly decreased (P < .001), K was significantly decreased (P < .001), α was significantly increased (P < .001), MA was significantly increased (P < .001), and there was no change in LY30 (P = .06).
|R (minutes)||1.1594 ± 0.19||1.5710 ± 0.43||<.0001|
|K (minutes)||1.0219 ± 0.42||2.3565 ± 1.32||<.0001|
|Αlpha (degrees)||75.728 ± 5.52||61.473 ± 9.88||<.0001|
|MA||68.109 ± 5.86||51.502 ± 6.39||<.0001|
|LY30||0.5469 ± 0.999||1.4742 ± 2.31||.06|
|ETP (nM/min)||424.07 ± 67.12||340.35 ± 63.09||.0008|
|Peak (nM)||155.29 ± 29.52||140.90 ± 19.94||.12|
|TTpeak (minutes)||3.0953 ± 0.48||2.7083 ± 0.30||.01|
|Lag (minutes)||1.3233 ± 0.28||1.1713 ± 0.20||.08|
Mean values for TG and standard deviations are presented in Table 1. When the HAC means were compared with the reference means, the values for ETP (P = .0008) were increased and TTpeak (P = .01) prolonged in the HAC population, whereas there was no difference seen in the means obtained for peak (P = .12) and lag time (P = .08).
As can be seen in Figure 4, there was no clear correlation between MA and ETP when individual dogs were compared.
A hypercoagulable TEG tracing is characterized by a decrease in the R and K values (signifying an acceleration in clot formation) as well as an increase in the α-angle and MA (associated with increased clot strength). When the HAC group was compared with the control group, a majority of patients (12/15) had evidence of hypercoagulability, implying that the prevalence of hypercoagulability in dogs with HAC is high, as has been previously reported. These findings are similar to those from a recent study that showed an increase in TEG MA in dogs with HAC.
However, few studies have shown a direct cause and effect relationship between a hypercoagulable TEG tracing and the development of thrombotic events, regardless of the underlying cause. No patients in our study had a clinically detectable thrombotic event and therefore this subset of patients could not be evaluated. We may not be recognizing thrombotic events in these animals, certain patients may have factors that protect them from developing thrombotic events, or TEG tracings may not be sufficiently sensitive in dogs for the detection of thrombosis resulting from HAC.
Finally, when evaluating the TEG values, the alpha value was abnormal in 10/15 dogs whereas the K value was abnormal in only 1/15 dogs. The K and alpha values normally are closely related. This observation could indicate that in these patients, the clot that formed was stronger and had more cross-linking, but did not form as rapidly as usual. Additional studies are needed to determine if this is the case.
CAT is a relatively new technique developed to measure the potential of thrombin generation. An increased ETP is a good marker of hypercoagulability. ETP recently has been used by our laboratory to evaluate the effect of heparin and prednisone on hemostasis in healthy dogs.
When our results are evaluated as a group, they support the TEG findings in that the study population of HAC dogs had increased ETP when compared with the reference interval. However, none of the other parameters (lag time, time to peak, and peak height) were compatible with increased thrombotic potential and, when the individual dogs were assessed, only 4 dogs showed hypercoagulable TG profiles. One dog even had a prolonged lag time that would be more suggestive of hypocoagulability. Furthermore, the individuals identified as hypercoagulable by TG profiles were not all the same as those identified by TEG profiles.
When the dogs were evaluated as a group, this study showed that TEG and TG both detected a hypercoagulable state in HAC patients reflected by an increase in MA and ETP, respectively. However, as shown in Fig 4, there does not appear to be a direct correlation between the 2 parameters.
The lower prevalence of hypercoagulability using the CAT assay (as compared to the TEG assay) may be explained by the protocol used. PPP was used instead of platelet-rich plasma for the measurement of TG. Therefore, the effect of platelets on TG was not assessed.
Unfortunately, platelet counts were only evaluated retrospectively and only counts that were low were evaluated by blood smear to determine if they were within normal limits. Nevertheless, a majority of dogs (11/14) had platelet counts above normal.
Our study had limitations, which included lack of age- and weight-matched control groups, lack of comparison with other established coagulation parameters (fibrinogen-degradation products, D-dimers, prothrombin time, activated partial thromboplastin time, dosage of coagulation factors), lack of assessment of effect of platelet counts, hematocrit and fibrinogen concentrations, and the mixture of both pituitary and adrenal HAC patients that were included in the study. Finally, the degree of hypercortisolemia was not evaluated, and this appears to contribute to the degree of hypercoagulability.
In conclusion, our study showed that both TEG and TG identified groups of dogs with HAC with a hypercoagulable profile, but identification of individual dogs is more difficult. Additional studies are needed to identify patients at risk for thromboembolism. Finally, additional factors that may influence coagulation should be assessed in these patients (eg, coagulation factors, platelet counts, and blood pressure).
Ania-Claude Lemaire and Martine Lamarre for their technical help; Dr Guy Beauchamp and Greg Johnson for the statistics; Dr Catherine Wagg, Dr Virginie Allegret, and Dr Carolyn Gara-Boivin for the laboratory analyses. This study was supported by the Association des Médecins Vétérinaires du Québec.
Conflict of Interest: Authors disclose no conflict of interest.
Vacutainer BD, Franklin Lakes, NJ
2.7 mL, 3.2% buffered sodium citrate (0.3 mL; 0.109M) Vacutainer, BD
Innovin, Dade Behring, Marburg, Germany
Thrombinoscope BV, Maatricht, The Netherlands
Fluoroskan Ascent, Thermolabsystems, Helsinki, Finland
Immulon 2 HD, Thermolab, Waltham, MA
SAS Institute, Cary, NC