In the current study, carcinomas located at other sites than the mammary glands were found to have an increased TEG G value, aPTT and fibrinogen compared to the other evaluated cancer types. In addition patients with distant metastatic disease were found to have elevated levels of fibrinogen and d-dimer compared to patients with local invasive and local non-invasive disease.
All the investigated cancer types in this study exhibited elevated median TEG G values and tended to be hypercoagulable, though the highest TEG G levels were seen in the group of patients with non mammary carcinomas ("carcinomas at other sites"). These patients were also found to have the highest median aPTT and fibrinogen level, even significantly when compared to the groups with mastocytomas and lymphomas.
Fibrinogen levels have previously been shown to positively correlate with TEG G in addition to platelet count [15, 22, 25]. We speculate, that the fibrinogen level in combination with the slightly elevated median aPTT and the haemostatic changes (elevated TEG G), likely is the net result of a systemic inflammatory reaction seen in these patients, in part mediated through proinflammatory changes in and around the cancer cells [26–28].
The patients with distant metastases were even found to have significantly increased fibrinogen levels compared to patients with local invasive and local non-invasive cancers, which is in accordance with findings made by other authors, who related hyperfibrinogenemia to stage of disease in canine mammary carcinoma patients . Also, increased fibrinogen levels have been associated with progression and extend of cancer disease in human patients . As previously mentioned the elevated fibrinogen seen in patients with metastatic disease indicates that systemic inflammation is associated with progression of the cancer disease.
The patients with distant metastatic disease had a significantly elevated d-dimer level and thereby more pronounced activation of the coagulation and fibrinolytic systems than patients with local invasive disease. Elevated d-dimer levels are found in both human and canine patients with VTE but also in other conditions such as wound healing, DIC and haemorrhagic conditions [29, 30]. None of the patients in this study had undergone surgery before blood sampling, but two of the patients with distant metastatic disease were diagnosed with DIC. None of the patients were diagnosed with VTE, though the presence of a thromboembolic event could have been missed. Due to decreased liver clearance, increased d-dimer can be found in patients with cirrhotic liver disease and two of the patients with distant metastasis had elevated liver enzymes. Anyway, it is unlikely that these patients, without clinical signs of liver involvement, were affected by their liver disease to an extend being the sole reason for alterations of their haemostatic and fibrinolytic profiles
Three of the included dogs in this study were found to be hypocoagulable on TEG. Two of the dogs had mammary carcinomas and metastases to either the regional lymph node or distant sites, while the third dog had lymphoma stage V. These findings are in accordance with previous findings by Kristensen et al. , where hypocoagulable dogs by TEG all had metastatic disease or lymphoma. The limited number of hypocoagulable dogs included in this study is not sufficient to state whether all hypocoagulable cancer patients have systemic cancer disease.
More than 90% of human cancer patients have subclinical haemostatic abnormalities with the majority of patients showing hypercoagulability associated with elevated fibrinogen levels, thrombocytosis, elevated levels of clotting factors V, VIII, IX and X and elevated levels of FDP, in addition to reduced levels of coagulation inhibitors (AT, Protein C and Protein S) and increased levels of plasminogen activator inhibitor-1 [1, 5, 31]. Similarly, the majority of canine cancer patients have been found to have subclinical haemostatic alterations as were found in this study [13–15, 19].
Haemostatic dysfunction and hypercoagulability in cancer patients is complex with a multifactorial aetiology that includes expression of procoagulant molecules such as tissue factor on the surface of malignant cells, release of fibrinolytic peptides, inhibition of endogenous anticoagulation, release of cytokines by cancer cells, and interaction with host cells including endothelial cells and blood leukocytes [28, 32]. Increased presence of activated platelets and release of platelet micro vesicles has also been related to hypercoagulability in cancer [33–35]. All these mechanisms additionally support cancer cell growth and metastasis.
In humans, hypercoagulability measured by TEG has been associated with increased risk of thromboembolism especially in post operative patients  but has not yet been identified as a predictor of VTE in cancer patients.
In order to verify the current findings, further studies including more patients of each cancer type and disease progression group should be performed. Furthermore the clinical impact of the hypercoagulable state in cancer patients and its possible association with VTE in both humans and dogs should be pursued. It appears that human and canine cancer patients develop similar haemostatic alterations, though thromboembolic complications are seldom recognized in dogs. The similarities of the haemostatic alterations, supports cancer bearing pet dogs as a valuable clinical translational animal model for studying the bidirectional link between haemostasis and cancer in humans.