**12.4 Chemotherapy**

Chemotherapy is one of the most important factors in VTE risk stratification of cancer patients. Large population-based studies in groups of pooled cancer patients have demonstrated a significantly increased risk in patients receiving chemotherapy. Heit et al used a population-based study of patients with a new diagnosis of VTE, 23% of which had a diagnosis of active malignancy, to demonstrate a significantly increased risk of VTE in those on chemotherapy (OR 6.5, CI 2.11–20) 146.

Studies in specific types of cancer and with specific antineoplastic agents have also supported the role of chemotherapy in predicting the risk of cancer-associated VTE. Two prospective studies of breast cancer patients demonstrated that the risk of VTE in patients receiving chemotherapy in addition to tamoxifen or surgery increased two- to seven-fold 147,148. A recent meta-analysis of breast cancer patients revealed that use of adjuvant hormonal therapy was associated with a 1.5–7-fold increased risk of VTE 149.

#### **12.5 Surgery**

12 Deep Vein Thrombosis

prior to diagnosis of cancer 135. In a population-based case–control study of patients with newly diagnosed VTE, including 389 patients with cancer, those with distant metastases had

A multicentre retrospective study of VTE in hospitalized cancer patients reported an incidence of 10.3% in patients with advanced-stage cancer compared with 5.6% in patients with localized disease (P < 0.0005, OR 1.92, CI 1.21–3.04) 137, and these findings have been supported by other large studies in hospitalized cancer patients 138. Other studies in ovarian, colorectal, pancreatic, lung and breast cancer support the finding that advanced-stage

In certain types of cancer, higher rates of VTE are found in some histological subtypes compared with others. For example, in patients with non-small-cell lung cancer, 9.9% of those with adenocarcinoma subtype develop VTE in the first 6 months after diagnosis compared with 7.7% with squamous cell carcinoma (HR 1.9, CI 1.7–2.1) 141.Inbreast and colon cancer patients, the type of histology does not predict for the incidence of cancerassociated VTE, but VTE-associated mortality rates are higher in patients with certain

Several studies have demonstrated that the risk of VTE is highest in the initial time period following cancer diagnosis. In a population-based study of patients with thrombosis, the risk of developing VTE was highest in the first few months following the initial diagnosis of malignancy. A retrospective analysis of over 200 000 cancer patients from the California Cancer Registry revealed that the rate of VTE per patient-year in the first year after diagnosis of cancer was 3.3, compared with 0.8 in the second year after diagnosis 145. The rate of VTE in patients with colon cancer during the first 6 months after diagnosis is 5.0/100 patient-years, but this drops off dramatically to 1.4/100 patient-years in the next 6- month

Chemotherapy is one of the most important factors in VTE risk stratification of cancer patients. Large population-based studies in groups of pooled cancer patients have demonstrated a significantly increased risk in patients receiving chemotherapy. Heit et al used a population-based study of patients with a new diagnosis of VTE, 23% of which had a diagnosis of active malignancy, to demonstrate a significantly increased risk of VTE in those

Studies in specific types of cancer and with specific antineoplastic agents have also supported the role of chemotherapy in predicting the risk of cancer-associated VTE. Two prospective studies of breast cancer patients demonstrated that the risk of VTE in patients receiving chemotherapy in addition to tamoxifen or surgery increased two- to seven-fold 147,148. A recent meta-analysis of breast cancer patients revealed that use of adjuvant

hormonal therapy was associated with a 1.5–7-fold increased risk of VTE 149.

a higher risk of VTE (OR 19.8, CI 2.6–149) 136.

**12.2 Histology** 

period 143.

**12.4 Chemotherapy** 

on chemotherapy (OR 6.5, CI 2.11–20) 146.

histological subtypes 141,143.

**12.3 Time after diagnosis** 

disease increases the risk of cancer- associated VTE 139,140,141,142,143,144.

Surgery is a well-known risk factor for development of VTE in patients without cancer. The incidence of DVT in cancer patients undergoing general surgery is estimated at 37% compared with 20% in patients without cancer 150. Factors related to immobility, tissue destruction and venous stasis are likely to be related to the increased risk of VTE after surgery.

#### **12.6 Indwelling catheters**

Indwelling central venous catheters (CVC) greatly facilitate treatment in cancer patients, but they are also associated with complications including a significant risk of catheter-associated thrombosis. The incidence of symptomatic catheter-related DVT in adult patients ranges from 0.3% to 28%, while the rate of catheter-related DVT assessed by venography is 27– 66%151.

Studies have not consistently demonstrated an association between use of haematopoietic growth factors and risk of cancer-associated VTE. İn a prospective study of ambulatory patients receiving chemotherapy, both the use of white cell growth factors and the use of red cell growth factors or decreased haemoglobin were independent predictors of VTE in multivariate analysis 152. This association was only significant in types of cancer already known to have high rates of thrombosis, and it is possible that these agents are used more frequently in patients with other markers of poor prognosis or more aggressive disease.

#### **12.7 Platelet and leukocyte counts**

The authors' group was the first to identify an elevated prechemotherapy platelet count as a significant risk factor for cancer-associated thrombosis 152. In a prospective study of outpatients receiving chemotherapy, 21.9% had a platelet count of 350 000/mm3 or more prior to starting chemotherapy. The incidence of VTE was 3.98% (1.66% per month) for these patients, which was significantly higher than the rate of 1.25% (0.52% per month) for patients with a prechemotherapy platelet count of less than 200 000/mm3 (P for trend¼ 0.0003). The distribution of rechemotherapy platelet counts in patients who subsequently developed VTE was significantly higher than that for patients who did not develop VTE (ttest P ¼0.002, Wilcoxon rank sum test P ¼0.0002).

#### **12.8 Tissue factor**

Tissue factor (TF), a transmembrane glycoprotein present on subendothelial tissue, platelets and leukocytes, is a key component in the initiation of coagulation and may play a role in cancer- associated thrombosis 153-155. The authors recently demonstrated a correlation between the level of TF expression in pancreatic tumours and subsequent development of VTE 156. VTE was four-fold more common (P ¼ 0.04) among patients with high TFexpressing carcinomas (26.5%) than among patients with low TF-expressing carcinomas (5.5%).

From 1979 to 1999, among 40,787,000 patients hospitalized in short-stay hospitals with any of 19 malignancies studied, 827,000 (2.0%) had VTE.157 This was twice the incidence in patients without these malignancies.157 The highest incidence of VTE was in patients with

Risk Factors of Deep Vein Thrombosis 15

women was higher than among white women.159,178,179 DVT was more frequent among women who underwent cesarean section (104/100,000/y) than those who underwent vaginal delivery (47/ 100,000/y).159 VTE in pregnancy is discussed in detail in the article by

In PIOPED, trauma of the lower extremities was a predisposing factor in 10% of patients with PE, and in PIOPED II trauma of the lower extremities or pelvis was a predisposing factor in 14%.180,181 Surgery within 3 months of the acute PE was a predisposing factor in 54% in PIOPED and in 23% in PIOPED II.180,181 The prevalence of VTE following various categories of surgery and trauma has been reviewed in detail by Geerts and colleagues182.

The use of long-term venous access is now an integral component of treatment for patients receiving long-term antibiotic administration or hyperalimentation or undergoing chemotherapy. Externalized tunneled catheters were introduced almost 30 years ago, but required daily cleaning and frequent flushing190,191. On average, deep venous thrombosis (DVT) can complicate approximately 2%–6.7% of such port placements192 – 194, although literature reports have ranged from 0% to 26%195–198. In 1991, Monreal et al.199 observed that 4 of 30 consecutive patients with upper extremity deep venous thrombosis (DVT) had PE (13.3%), but more importantly, all these 4 occurred in 20 catheter related DVT patients

The incidence of VTE among hospitalized medical patients with ulcerative colitis was 1.9% and the incidence with Crohn disease was lower (1.2%).200 Among medical patients who had neither ulcerative colitis nor Crohn disease the incidence was 1.1%.200 The relative risk of VTE among patients with ulcerative colitis compared with patients who did not have inflammatory bowel disease was 1.9 and with Crohn disease it was 1.2. Among patients younger than 40 years with ulcerative colitis, the relative risk of VTE compared with patients who did not haveinflammatory bowel disease was 2.96 and in patients younger

Patients with chronic liver disease (both alcoholic and nonalcoholic) seem to have a lower risk of PE than patients without liver disease,43,201 but data are inconsistent.202 Chronic liver disease may result in impaired production of vitamin-K dependent procoagulant factors.203 However, decreased production of vitamin-K dependent endogenous anticoagulants, such as protein C, protein S, and antithrombin III, may counter the hypocoagulability in such patients.203 Other prothrombotic factors may counteract the impaired production of vitamin Kdependent procoagulant factors including lupus anticoagulant, activated protein C resistance, PT20210A mutation, Factor V Leiden, MTHFR

(20%), while none of 10 patients with primary upper extremity DVT had PE.

than 40 years with Crohn disease the relative risk was 2.23.200

Marik elsewhere in this issue.

**14. Surgery and trauma** 

**15. Central venous access** 

**16. Medical illnesses** 

**16.2 Liver disease** 

**16.1 Inflammatory bowel disease** 

carcinoma of the pancreas (4.3%) and the lowest incidences were in patients with carcinoma of the bladder and carcinoma of the lip, oral cavity, or pharynx (<0.6% to 1.0%). Incidences with cancer were not age dependent.157 Myeloproliferative diseaseand lymphoma were associated with relative risks for VTE of 2.9 and 2.5, respectively157 Leukemia was associated with a lower relative risk (1.7). Based on death certificates from 1980 to 1998 among patients who died with cancer, PE was the listed cause of death in 0.21%.158 Adjustment of the data for the frailty of the diagnosis of fatal PE based on death certificates indicated a likely range of 0.31% to 1.97%.158
