Assessment of Thrombotic Risk in Cancer Patients: Proposal of a Scoring System. A Monocentric Study.

Background Deep venous thrombosis (DVT) is present in about 3–15% of cancer patient. Paraneoplastic thrombosis pathogenesis is multifactorial. Antithrombotic prophylaxis frequently is not oriented to the real thrombotic or haemorragic risk of the patients. Aim Aim of our study is to define the real thrombotic risk in neoplastic patient. With this purpose we considered in our patients complement fraction C3 and C4 and circulating immune-complexes (CIC) because they activate macrophages and platelets and increase tissue factor level. Moreover we evaluated also total cholesterol and triglycerides level, because they are linked with factor VII activation. Chemotherapy dose density and chemotherapeutic agents used were also included in our investigation as pro-inflammatory/pro-thrombotic risk factors.

Methods We considered C3, C4, CIC, cholesterol and triglycerides level, dose density and chemotherapeutic agents used in 119 patients with solid neoplasms (62 colon, 27 lung, 18 gastric, 12 others) and without anticoagulant prophilaxis. Of these only 105 were evaluable because in these all the data were available. Median age was 68.5 years (R 57–83). M/F ratio was 72/47. The threshold value of third quartile was chosen as risk cut-off (C3: 130mg/dl; C4: 32mg/dl; ICC: 2.9 mcg/ml; total cholesterol: 205mg/dl; triglycerides 123mg/dl). We elaborated a scoring system in which 1 point was attributed to each value inferior to the third quartile. One point was added or removed if patient received or not Irinotecan. In fact this drug was the only variable that in univariate analysis showed an high Yates corrected chi square test (4, p 0.045). The statistical analysis was conducted with Yates corrected chi square test, Odds Ratio (OR), realtive risk (RR). Furthermore Erythrocyte Sedimentation Rate (ESR) and C-reactive Protein (CRP) were also evaluated.

Results 27 patients (26%) showed DVT. Of these 23(85%) had a score ≤3 and 4 (15%) ≥4. 78 patients (74%) did not show DVT. Of these 46(59%) had a score ≥4 and 26 ≤3. Yates corrected chi square test is 15.68 (p <0.0001), with an OR of 8.2 (95%CI 2.7–24.9) and a RR of 5.2 (95%CI 2.1–13.9). Negative predictive value is 0.92 (95%CI 0.83–0.96) and positive predictive value is 0.42 (95%CI 0.34–0.46). Sensitivity was 0.85 (95%CI 0.70–0.93) and specificity was 0.59 (95%CI 0.53–0.62). Duration of chemotherapy was not influent on DVT onset (Chi-Square 0.8, p 0.3; OR 0.6). CRP and ESR at Pearson’s test were not related with C3, C4, CIC, total cholesterol and triglyceride levels. Colon and stomach cancer were frequently related with DVT onset (62% of all DVT cases).

Summary/Conclusions The neoplastic patient frequently shows haemorrhagic risk (eg in chemotherapy induced thrombocytopenia). Therefore anti-thrombotic prophylaxis should be used considering the real thrombotic and haemorragic risk of neoplastic patient. Our scoring system is useful in distinguishing cancer patient with low thrombotic risk and consent to avoid antithrombotic prophylaxis in patient with higher bleeding and lower thombotic risk. Nevertheless these data need confirmation on a larger cohort of patients.