Introduction

Hemostatic dysfunction is common in both human and veterinary oncologic patients.^1–5 Human cancer patients often present with clinical signs of venous thromboembolism or pulmonary thromboembolism.¹ In addition, human patients with malignancy have higher risk of developing thromboembolism after surgeries.⁶ Pulmonary thromboembolism has also been reported in dogs with neoplasia.⁷ Thrombosis of the portal vein, aorta, and iliac and splenic arteries has also been reported in dogs with neoplasia.^8–13

Thromboelastography (TEG) is a whole blood assay that yields global assessment of hemostasis, and it evaluates clot time, strength, and kinetics of clot formation and lysis. It is a point-of-care test that has been validated in dogs, cats, and horses to detect hyper-, hypo-, and normocoagulable states. Recent studies^2–4 have demonstrated hemostatic disorders including hypercoagulability in dogs with neoplasia.

A hypercoagulable state has been reported in both human and dogs with hyperadrenocorticism (HAC). Literature reviews have suggested a high risk for venous thromboembolism in human patients with HAC.^14–17 Dogs with HAC have also been reported to have evidence of hypercoagulability based on TEG results in multiple studies.^18–20 However, conflicting results were reported in 1 study,²¹ which did not support a significant difference between coagulation tendencies between dogs with HAC, dogs that were treated for HAC, and dogs without HAC. Thrombosis of the caudal vena cava, portal vein, abdominal aorta, iliac arteries, and pulmonary arteries has been reported in dogs with HAC or dogs with iatrogenic HAC.^7,8,10,22,23 Given these findings, it is possible that dogs with adrenal tumors receiving adrenalectomy may be in a hypercoagulable state, and this may pose a higher risk for developing postoperative thromboembolic disease.

To the best of our knowledge, preoperative TEG findings in dogs undergoing adrenalectomy have not been reported. The objective of this study was to describe TEG findings in dogs that had adrenalectomy performed. In addition, we aimed to compare the TEG findings for dogs with and without HAC receiving adrenalectomy.

Methods

Results

The medical records search identified 35 dogs that had adrenalectomy in the study period. One dog was excluded because TEG was run but results were not available, 2 dogs were excluded because of missing surgical or histopathology reports, and 2 dogs were excluded because of administration of anticoagulants within 1 month prior to surgery.

For the 30 dogs included in the study, the median age at the time of the surgery was 11 years old (range, 1 to 14 years) and the median body weight was 17.7 kg (range, 2.4 to 57.1 kg). There were 14 spayed females, 15 neutered males, and 1 sexually intact male. The most common clinical sign at presentation was polyuria/polydipsia (n = 13), followed by panting (8), and lethargy (7). Other clinical signs included alopecia (n = 5), polyphagia (4), weight loss (3), hyporexia (3), abdominal distension (3), and vomiting (1). For preoperative imaging and staging, 17 dogs had abdominal CT, 4 had abdominal ultrasound, and 9 had both abdominal CT and abdominal ultrasound. Fourteen dogs had thoracic radiographs, 15 had thoracic CT, and 1 dog had no record of thoracic imaging.

Based on G value, 16 of 30 (53.3% [95% CI, 35.5% to 71.1%]) of the dogs were hypercoagulable and 14 of 30 (46.7% [95% CI, 28.3% to 65.7%]) were normocoagulable (example of TEG traces in Figure 1A & B). None of the dogs were hypocoagulable. A summary of preoperative TEG findings in normocoagulable and hypercoagulable dogs is presented (Table 1). A CBC was available in 29 dogs, with 6 of 29 (20.7% [95% CI: 8.0-39.7%]) dogs having thrombocytosis, and of these, 3 out of 6 dogs were hypercoagulable based on TEG G values. Alanine aminotransferase and ALP values were available in 29 dogs; ALT was elevated in 14 of 29 (41.4% [95% CI, 30.1% to 66.5%]), ALP in 22 of 29 (75.9% [95% CI, 56.5% to 89.7]), and AST in 6 of 28 (21.4% [95% CI, 8.3% to 41.0%]) of dogs. Six out of 14 (95% CI, 17.7% to 71.1%) dogs with elevated ALT were hypercoagulable, and 13 of 22 (59.0% [95% CI, 36.4% to 79.3%]) dogs with elevated ALP were hypercoagulable. Total protein was elevated in 22 of 29 (75.9% [95% CI, 56.5% to 89.7%]) of dogs, and of these, 12 of 22 (54.5% [95% CI, 32.2% to 75.6%]) were hypercoagulable based on TEG G values. The PT and aPTT were available in 5 dogs and were within normal range in all 5 dogs; however, 4 of 5 (95% CI, 28.4% to 99.5%) of these dogs were identified as hypercoagulable based on TEG G values. Fibrinogen level was available and elevated in only 1 dog, which was considered hypercoagulable based on TEG G value.

View Full Size

View Full Size

Example of thromboelastographic (TEG) tracings in dogs. A—Normal tracing. The x-axis represents time in minutes and the y-axis the amplitude of pin movement in millimeters. Reaction time (R) is from the start of the test to the first detection of fibrin clot formation and when the amplitude reaches 2 mm. Clot time (K) is the time from clot formation (end of R) until amplitude of 20 mm is reached. α Angle (Angle) is the angle tangent to the curve as K is reached and represents the acceleration of fibrin formation and cross-linking. Maximum amplitude (MA) is the maximum amplitude of the TEG curve measured in millimeters and reflects the maximum clot strength or stiffness. B—Hypercoagulable tracing. Note the increase in α angle and in MA, giving the tracing a “wineglass” appearance.

Citation: Journal of the American Veterinary Medical Association 262, 6; 10.2460/javma.23.08.0456

• Download Figure
• Download Figure
• Download figure as PowerPoint slide

Fourteen out of 30 (47%) dogs were tested for HAC, with 8 (27%) of those dogs diagnosed with HAC based on consistent clinical signs and physical examination abnormalities in addition to consistent endocrine testing results. An ACTH stimulation test was performed in 3 dogs, a low-dose dexamethasone suppression test was performed in 9 dogs, and both tests were performed in 1 dog. Four out of the 8 dogs with HAC were hypercoagulable. Three HAC dogs were on trilostane, and 2 of 3 of these dogs were hypercoagulable. There were no other preoperative medications used for treatment of HAC.

Right adrenalectomy was performed in 16 of 30 (53%) dogs, left adrenalectomy was performed in 11 of 30 dogs (37%), and bilateral adrenalectomy was performed in 3 of 30 (10%) dogs. Vascular invasion was present in 4 of 30 adrenalectomies (13%; 3 right and 1 left), and of these, a vena cavotomy was performed in 2 of 4 (1 right, 1 left); in the remaining 2, the vascular invasion extended only into the phrenicoabdominal vein. Ligation/transection of the phrenicoabdominal vein was sufficient for the removal of the adrenal gland.

Hemorrhage was the most common intraoperative complication (n = 5). One dog had a severe hemorrhage intraoperatively, which resulted in hypovolemic shock, hypoxia, bradycardia, and near cardiac arrest. After multiple blood transfusions, the dog appeared to recover; however, the dog developed pneumothorax and acute hemoabdomen postoperatively and was later euthanized. Other intraoperative complications included hypotension (n = 2) and hypothermia (1). The most common additional procedure was liver biopsy (9/30 [30%]). Other procedures included lipoma excision (n = 3), gastropexy (2), and 1 each of tongue biopsy, soft tissue sarcoma excision, mast cell tumor excision, anal sacculectomy for apocrine gland adenocarcinoma, cholecystectomy, colopexy, and cystopexy.

Ten (33%) dogs were diagnosed with adrenocortical adenoma, 9 (30%) with adrenocortical adenocarcinoma (ACA), and 8 (27%) with pheochromocytoma; 1 (3%) dog was diagnosed with adrenocortical hyperplasia, 1 (3%) with poorly differentiated sarcoma, and 1 (3%) with both ACA and pheochromocytoma. In the dog diagnosed with both ACA and pheochromocytoma, the dog had unilateral adrenalectomy, and the 2 different neoplasia types were found in 1 adrenal gland. No dogs with ACA or pheochromocytoma that had biopsies performed intraoperatively had evidence of metastatic disease. Six of 9 (95% CI, 29.9% to 92.5%]) dogs with ACA, 4 of 8 (95% CI, 15.7% to 84.3%]) with pheochromocytoma, and 6 of 10 (95% CI, 26.2% to 87.8%]) with adenoma were hypercoagulable (P = .28; Fisher exact test). None of the 3 dogs with other histopathologic diagnoses or combinations of diagnoses (adrenocortical hyperplasia, poorly differentiated sarcoma, and the 1 dog diagnosed with both ACA and pheochromocytoma) were hypercoagulable.

Discussion

To our knowledge, this is the first time preoperative TEG findings for dogs undergoing adrenalectomy have been reported. Overall, approximately half of the dogs having adrenalectomy in this study were hypercoagulable, and none were hypocoagulable. Hypercoagulability was seen in most dogs with ACA and adenoma and in 50% of dogs with pheochromocytoma. There was no significant difference in hypercoagulability between the tumor types. This study’s major findings were similar to those of other studies, including where hypercoagulability was also one of the most common findings in dogs with neoplasia compared to controls.^3,4 However when Appelgrein et al²⁴ evaluated hemostatic abnormalities with rotational TEG (ROTEM) in dogs undergoing adrenalectomy, they found lower proportions of dogs with hypercoagulability. Overall, in that study,²⁴ in which 25 of the 38 (66%) dogs undergoing adrenalectomy were tested for ROTEM, 21.1% (8/38) of dogs that had adrenalectomy were hypercoagulable, 18.4% (7/38) of dogs were hypocoagulable, and 60.5% (23/38) of dogs were normocoagulable. This is different from our study findings of 55%, 0%, and 45%, respectively, and it is markedly lower than 50% (18/36) of the malignant neoplasia reported as hypercoagulable by Kirstensen et al.⁴ The variability of the proportion of hypercoagulable dogs between the studies is difficult to explain, and there may be a few possible reasons. Firstly, both TEG and ROTEM are considered global assays of hemostasis, as they are viscoelastic assays that evaluate cellular (platelet) and protein (coagulation factors) components of hemostasis and fibrinolysis. They both function on the same principle, which measures the movement between the pin and the cup as the clot forms. The main differences are that they have different nomenclature and whether the pin rotates relative to the cup in the case of ROTEM or vice versa in TEG.^25–27 The difference in pin/cup rotation that determines the clot formation may potentially contribute to the difference in the results. Differences in viscoelastic testing results can occur between institutions and through the use of different activators.²⁵ The parameters used to define hypercoagulability could possibly contribute to the difference in the results. If R and K values were used, this can potentially contribute to greater difference in results, as they are more sensitive to the activator type used.²⁸ Finally, there is the question of validity of comparing the results of TEG and ROTEM studies themselves.

Despite the limited numbers of dogs tested for HAC, half of the dogs diagnosed with HAC were found to be hypercoagulable in this study. Three of the 8 dogs diagnosed with HAC were on trilostane preoperatively. However, 2 out of these 3 dogs were hypercoagulable despite being on trilostane. Follow-up diagnostic results to evaluate whether HAC was well controlled with trilostane therapy were not available, and it is possible that these 2 dogs could still have had uncontrolled HAC. However, it may also be that these dogs were well controlled after treatment, yet still remain hypercoagulable, as this has been documented in other studies.^{18, 20}

In this study, hypocoagulabilty was not found in any dogs compared to prior studies^2,4 where hypocoagulability was found in dogs with lymphoma or metastatic disease. However, other studies^3,29 did not identify hypocoagulabilty despite dogs having regional or distant metastases. In our study, none of the dogs were hypocoagulable, and none had metastasis. One possible reason that none of the dogs in our study had metastasis may be selection bias, with those found with metastasis during staging perhaps not having been encouraged to undergo surgery due to a worse prognosis.

There are some limitations of this study that should be considered when interpreting the results. These include the retrospective nature of the study, which resulted in a lack of standardization protocol in determining which dogs were tested for HAC, and it is possible some of the dogs that were not tested may have had HAC. In addition, there were inconsistencies in pretreatment with trilostane for those diagnosed with HAC and in the use of postoperative treatment with clopidogrel. In addition, fibrinogen was evaluated in only 1 dog, a parameter that is considered when reporting TEG results. Proteinuria, which could have contributed to hemostatic state, as it is not uncommon for HAC dogs to be proteinuric, was not assessed. Also, 3 dogs had 1 each of soft tissue sarcoma, mast cell tumor, and apocrine gland adenocarcinoma. The presence of these different neoplastic conditions may have had an effect on the TEG results. Lastly, low sample size (n = 30) in this case series often resulted in wide 95% CIs, particularly when data were stratified by tumor type.

In conclusion, we diagnosed hypercoagulability based on G values in TEG in dogs with carcinoma, pheochromocytoma, and adenoma. This was a retrospective study, and not all patients had a complete endocrine workup performed including testing for HAC where relevant. Endocrine testing was at the discretion of the primary clinician. Eight of 14 dogs tested were diagnosed with HAC, and only 4 of 8 dogs with HAC were hypercoagulable. It may be that dogs with adrenal tumors and HAC have a higher risk of hypercoagulability than dogs with no HAC and other adrenal tumors. Further prospective study, with standard endocrine testing for HAC, would be needed to test this hypothesis. In addition, prospective studies are needed to determine whether a hypercoagulable state identified by TEG correlates with clinical development of thromboembolic disease and whether HAC dogs are at greater risk of developing TEG complications.

Disclosures

The authors have nothing to disclose. No AI-assisted technologies were used in the generation of this manuscript.

Funding

The authors have nothing to disclose.