Presence of varicose veins in cancer patients increases the risk for occurrence of venous thromboembolism

Transcription

1 Journal of Thrombosis and Haemostasis, 11: DOI: /jth ORIGINAL ARTICLE Presence of varicose veins in cancer patients increases the risk for occurrence of venous thromboembolism O. K ON I G S B R UG G E, * F. L OTSCH,* E.-M. REITTER,* T. BRODOWICZ, C. ZIELINSKI, I. PABINGER* and C. A Y * *Clinical Division of Hematology and Hemostaseology, Department of Medicine I, Medical University of Vienna; and Clinical Division of Oncology, Department of Medicine I, Medical University of Vienna, Austria all Comprehensive Cancer Center Vienna all Medical University of Vienna, Vienna, Austria To cite this article: K onigsbr ugge O, L otsch F, Reitter E-M, Brodowicz T, Zielinski C, Pabinger I, Ay C. Presence of varicose veins in cancer patients increases the risk for occurrence of venous thromboembolism. J Thromb Haemost 2013; 11: Summary. Background: Cancer patients are at increased risk of venous thromboembolism (VTE). Objective: We investigated the association of a history of VTE, superficial thrombophlebitis, or the presence of varicose veins with the occurrence of VTE during the course of cancer. Methods: Cancer patients were recruited in a prospective cohort study, the Vienna Cancer and Thrombosis Study. Patients who had VTE within 3 months before study inclusion were excluded. At study inclusion, history of VTE, history of superficial thrombophlebitis, and presence of varicose veins were recorded. Primary end point was the occurrence of symptomatic VTE. Hazard ratios were obtained using the competing risk analysis according to Fine and Gray. Results: The cohort consisted of 1270 patients followed over a median of 590 days. A history of VTE was found in 66 patients (5.2%), superficial thrombophlebitis in 79 patients (6.2%), and varicose veins in 160 patients (12.6%). Ninety-eight patients (7.7%) developed VTE during follow-up. The hazard ratios for the risk of VTE in patients with a history of VTE or superficial thrombophlebitis were 1.44 (95% confidence interval: ) and 1.94 ( ), respectively, and 2.01 ( ) in those with varicose veins. In multivariable analysis including history of VTE, history of superficial thrombophlebitis, presence of varicose veins, and other patient-related factors, the presence of varicose veins (2.10 [ ]) remained significantly associated with an increased risk of VTE. Conclusion: The presence of Correspondence: Cihan Ay, Clinical Division of Hematology and Hemostaseology, Department of Medicine I, Comprehensive Cancer Center, Medical University of Vienna, Waehringer Guertel 18-20, A Vienna, Austria. Tel.: ; fax: cihan.ay@meduniwien.ac.at Received 4 June 2013 Manuscript handled by: M. Cushman Final decision: F. R. Rosendaal, 12 September 2013 varicose veins is associated with an elevated risk of VTE in cancer patients. This clinical parameter could be useful for individual risk assessment of VTE in these patients. Keywords: neoplasm; risk; thrombophlebitis; varicose veins; venous thromboembolism. Introduction The association of cancer and venous thromboembolism (VTE) is well established. Since VTE events in cancer patients are accompanied with high mortality and morbidity [1 4], further research into the details of the association between cancer and VTE is necessary. In the general population, many risk factors for VTE have been identified. Superficial thrombophlebitis, varicose veins, and a history of VTE in particular increase the risk of VTE [1,5,6]. In cancer patients, several patient-specific risk factors for cancer-associated VTE have been described [7,8]. Similar to the general population, a history of previous VTE has been regarded as a risk factor for recurrent VTE events in cancer patients. For instance, Agnelli et al. reported that surgical cancer patients with a history of VTE had a six-fold increased risk of new VTE events after surgery [9]. Similarly, a four-fold increased VTE risk was found for multiple myeloma patients and a 20-fold increased VTE risk was found for patients with ovarian cancer and a history of VTE [10,11]. In another prospective observational study on a cohort of medical patients with different cancer entities, a two-fold increase in VTE events was found in patients with history of VTE [8]. Prandoni et al. further demonstrated that cancer patients are still at higher risk of recurrent VTE compared with non-cancer patients, even while receiving anticoagulation therapy [12]. However, it has not been entirely elucidated whether a history of previous VTE that did not occur in close

2 1994 O. K onigsbr ugge et al temporal proximity to the diagnosis of cancer and was probably not cancer related is associated with an elevated risk of VTE during the course of active cancer. Likewise, it has not been explicitly studied whether a history of superficial venous thrombophlebitis or the presence of varicose veins at cancer diagnosis also contribute VTE risk in cancer patients. Hence, in a prospective observational cohort study of cancer patients, the Vienna Cancer and Thrombosis Study (CATS), we investigated the association of a previous history of VTE and superficial thrombophlebitis and the presence of varicose veins with the risk of developing VTE during the course of cancer. Methods Study design and patient population The CATS is an ongoing prospective cohort study aimed at investigating predictive parameters for the occurrence of VTE in cancer patients [13 15]. The study is performed at the Vienna General Hospital of the Medical University of Vienna, Austria, in accordance with the Declaration of Helsinki and has been approved by the local ethics committee. The study s inclusion criteria were (i) newly diagnosed, active cancer of the brain, breast, lung, kidney, prostate, gastrointestinal tract, pancreas, or other solid tumors (mainly gynecological tumor or sarcoma) or hematologic malignancies or progression of disease after complete or partial remission; (ii) histologic confirmation of cancer diagnosis; (iii) age of at least 18 years; (iv) willingness to participate; and (v) written documentation of informed consent. All inclusion criteria needed to be met. Patients were excluded according to the following criteria: (i) acute bacterial or viral infection within the past 2 weeks; (ii) continuous anticoagulation therapy; (iii) chemotherapy within the past 3 months; (iv) radiation therapy; or (v) major surgery within the past 2 weeks; and (vi) VTE or arterial thromboembolism within the past 3 months prior to the time of inclusion. Patients who recently received chemotherapy, radiotherapy, or surgery before study inclusion were excluded due to possible transient effects of these treatments to the hemostatic system. Patients were allowed medication with aspirin or clopidogrel and temporary anticoagulation with low molecular weight heparin during hospital stays, which required thromboprophylaxis. On study admission, patients underwent a structured interview by a physician, trained in recruiting patients for the study with a questionnaire about their medical history, family history, long-term medication, and comorbidities. Information on the type of malignancy and cancer therapy was also recorded. Patients were given detailed information on the symptoms of VTE (deep vein thrombosis [DVT] and/or pulmonary embolism [PE]) and how to contact the study administration in case of an event. During the course of the 2-year observation period, the patients were contacted approximately every 3 months via mail questionnaire regarding possible VTE events, their current medical status, cancer progression, and changes in therapy. If patients did not respond to mail, their family members, attending oncologists, and general practitioners were contacted. Once a year, the Austrian death registry was searched for patients who had been included in the study. If patients omitted replying to single mail questionnaires, they were not treated as lost to follow-up if overall contact was maintained. Of the 1636 patients originally recruited from October 2003 until May 2011, 195 were excluded after revaluation, for not meeting all inclusion criteria. An additional 171 patients could not be included in this analysis for lack of follow-up reports. For the current analyses, complete data were available in 1270 subjects. Outcome measurement The primary end point of the study was the occurrence of a symptomatic or fatal VTE event within an observation period of 2 years. When patients developed symptoms of VTE, they were subjected to diagnosis with imaging methods as previously described [13,15]. DVT had to be confirmed with either duplex sonography or venography. Patients with PE were diagnosed using computed tomography or lung perfusion scanning. The VTE events had then to be confirmed by an adjudication committee, a board of independent experts from the fields of radiology, angiology, and nuclear medicine. Ascertainment of history of VTE and superficial thrombophlebitis and presence of varicose veins During the structured interview at study inclusion, a history of previous VTE was questioned in detail for location of occurrence, symptoms, imaging methods performed, medical diagnosis, and subsequent treatment to distinguish between a history of DVT and a history of superficial thrombophlebitis. Patients were specifically asked about swelling, heaviness, cramping, itching, painful induration, and color changes of a superficial vein, as well as whether treatment with low molecular weight heparin had been initiated. Patients were directly questioned by a study physician about the presence of varicose veins. The diagnosis for varicose veins was based on clinical inspection according to CEAP (Clinical condition, Etiology, Anatomic location and Pathophysiology) Classification. If study physicians discovered visible or palpable varicose veins, they were considered to be CEAP score 2 or higher and clinically significant [16]. In addition to the information gained from the structured interview on the history of VTE and superficial thrombophlebitis and the presence of varicose veins, the

3 Venous diseases and risk of cancer-associated VTE 1995 records of telephone conversations with patients, family members, and family practitioners were consulted. Finally, the Vienna General Hospital s digital information system was retrospectively searched to include all medical records from 3 months before study inclusion until 1 month after study inclusion, for the diagnosis of previous VTE or superficial thrombophlebitis or presence of varicose veins. The main characteristics of the study population, as well as the main characteristics of patients with VTE events during the observation time, are shown in Table 1. Patients were prospectively followed for a median time of 590 days (25th 75th percentile: ). Four hundred ten patients (32.3%) died during follow-up. The median time to death was 249 days (25th 75th percentile: ). Statistical analysis The continuous variables age, height, and weight were described using median and 25th 75th percentiles. Categorical variables were described with absolute and relative frequencies. The median follow-up time was calculated with the reverse Kaplan Meier method. New VTE events during the observation period were described with their absolute frequencies and with the cumulative probability of VTE occurrence after 180 and 365 days using the cumulative incidence function for competing risk. The group of patients with VTE events during the observation period was further described with the characteristics of age, sex, body mass index, site of cancer, location of VTE event, history of VTE, history of superficial thrombophlebitis, and presence of varicose veins. The risk of VTE for patients with prior history of VTE and of superficial thrombophlebitis and presence of varicose veins was calculated with univariate and multivariable proportional subdistribution hazards models for competing risk according to Fine and Gray [17]. The implementation of the competing risk analysis is recommended by the Scientific Subcommittee (SSC) on Haemostasis and Malignancy of the International Society on Thrombosis and Haemostasis (ISTH) in studies with competing risk of death with VTE [18]. A first multivariable model included history of superficial thrombophlebitis and history of VTE and a second model in search of confounders was adjusted for history of VTE, history of superficial thrombophlebitis, and varicose veins (Model 2, Table 3). Both multivariable competing risk regression models were adjusted for the variables age, gender, body mass index, and whether cancer was newly diagnosed or in progression. Due to the exploratory character of this study, a significantly elevated risk was considered for a P- value < All calculations were performed with Stata (Version 12.0 for Windows, Stata Corp., College Station, TX, USA). Results Study population The study population included 1270 cancer patients, of whom 716 were male (56.4%). The median age of study participants was 62 years (25th 75th percentile: 52 69). VTE events During the follow-up period, 98 patients (7.7% of study population) developed a VTE event. Characteristics of the patients who developed VTE are described in Table 1. The respective sites of VTE events are listed in Table 2. The cumulative probability for occurrence of VTE was 6.1% after 180 days and 7.8% after 365 days of study inclusion. Table 1 Characteristics of total study population and patients with VTE events during observation time Total study population characteristics, N = 1270* Patients with VTE events during observation time, N = 98 Sex, N (%) Female 554 (43.6) 37 (37.8) Male 716 (56.4) 61 (62.2) Age at study entry, 62 ( ) 61 ( ) years [median (IQR)] Age at event, years n.a ( ) (median [IQR]) BMI, kg m ² 25.2 ( ) 25.2 ( ) (median [IQR]) Site of cancer, n (%) Brain 150 (11.8) 24 (24.5) Breast 168 (13.2) 3 (3.1) Lung 187 (14.7) 13 (13.3) Stomach 53 (4.2) 8 (8.2) Colon 135 (10.6) 12 (12.2) Pancreas 78 (6.1) 15 (15.3) Kidney 39 (3.1) 1 (1.0) Prostate 138 (10.9) 3 (3.1) Multiple myeloma 33 (2.6) 2 (2.0) Lymphoma 203 (16.0) 12 (12.2) Other cancer entity 86 (6.8) 5 (5.1) History of VTE, n (%) 66 (5.2) 7 (7.1) Isolated DVT of the 47 (3.7) 4 (4.1) lower extremity Isolated PE 15 (1.2) 2 (2.0) Combined DVT of the 4 (0.3) 1 (1.0) lower extremity and PE History of superficial 79 (6.2) 11 (11.2) thrombophlebitis, n (%) Varicose veins, n (%) 160 (12.6) 22 (22.4) VTE indicates venous thromboembolism; DVT, deep vein thrombosis; PE, pulmonary embolism; IQR, interquartile range; BMI, body mass index. *Number in brackets denotes the percentage of the entire study population. Number in brackets denotes the percentage of patients with VTE events during the observation period.

4 1996 O. K onigsbr ugge et al Table 2 Characteristics of patients who developed VTE during follow-up Site of VTE, N (% of total number of VTE events) Isolated DVT of the lower extremity 41 (41.8%) Distal DVT 6 (6.1%) Proximal DVT 35 (35.7%) Isolated PE 34 (34.7%) Combined PE and DVT of the lower extremity 5 (5.1%) Portal vein thrombosis 4 (4.1%) Sinus vein thrombosis 1 (1.0%) Fatal PE 4 (4.1%) Combined DVT of the lower extremity 1 (1.0%) and the portal vein Combined PE and DVT of the upper extremity 1 (1.0%) DVT of the upper extremity 3 (3.1%) Inferior caval vein thrombosis 1 (1.0%) Internal jugular vein thrombosis 3 (3.1%) VTE indicates venous thromboembolism; DVT, deep vein thrombosis; PE, pulmonary embolism. History of VTE and association with the occurrence of cancer-associated VTE A history of previous VTE that had occurred at least 3 months before study inclusion was reported in 66 patients (5.2%). The index event was DVT in 51 cases and PE in 19 cases. Four patients had previously had both DVT and PE. New VTE events during the observation period occurred in 91 patients (7.6%) without history of VTE and 7 patients (10.6%) with history of VTE. The cumulative probabilities for the occurrence of VTE were 6.1% for patients without a history of VTE and 5.1% for those with a history of VTE during the first 180 days after study inclusion. After 365 days, the cumulative probabilities were 7.5% in patients without and 13.6% in those with a history of VTE (Fig. 1). In univariate competing risk regression analysis, no significantly increased risk for VTE was found for patients with previous VTE events (hazard ratio [HR] = 1.44, 95% confidence interval [CI] ; P = 0.347). In the multivariable proportional subhazards model, this association was statistically not significant (HR = 1.36, 95% CI ; P = 0.433) (Table 3) after adjusting for varicose veins, history of superficial thrombophlebitis, newly diagnosed cancer/progression of cancer, sex, age, and obesity. History of superficial thrombophlebitis and association with cancer-associated VTE We identified 79 patients (6.2%) with a history of superficial thrombophlebitis. During follow-up of the study, VTE events occurred in 11 patients (13.9%) with a history of superficial thrombophlebitis compared with 87 patients (6.8%) without a history of superficial thrombophlebitis. The cumulative probabilities for VTE occurrence were 5.7% for patients without and 10.6% for patients with a history of superficial thrombophlebitis after 180 days, and 7.3% and 13.9% after 365 days (Fig. 2). In univariate competing risk regression analysis, the risk of VTE was significantly increased in patients with a history of superficial thrombophlebitis (HR = 1.94, 95% CI ; P = 0.038). In the multivariable proportional subhazards model including newly diagnosed cancer/progression of cancer, gender, age, obesity, and prior history of VTE, a history of superficial thrombophlebitis showed a statistically significant association with future VTE (HR = 1.97, 95% CI ; P = 0.038). Of the 79 patients with a history of superficial thrombophlebitis, 18 also had preexisting varicose veins. Four VTE events occurred among patients with both a history of superficial thrombophlebitis and varicose veins. Presence of varicose veins and a history of superficial thrombophlebitis were Cumulative probability of VTE (%) Follow-up time (days) No prior history of VTE Prior history of VTE Fig. 1. Cumulative incidence of VTE. Patients with history of VTE are compared to patients without history of VTE. Table 3 Multivariable proportional hazards model for VTE occurrence in cancer patients (N = 1270) Parameters HR CI P Model 2 Varicose veins History of superficial thrombophlebitis History of VTE Gender* Age BMI New cancer diagnosis/progression VTE indicates venous thromboembolism; HR, hazard ratio; CI, confidence interval; BMI, body mass index. *Parameter gender shows the hazard of female sex. Parameter age shows an increase in hazard for every 1 year added to age. Parameter BMI shows an increase in hazard for every 1 kg m 2added.

5 Venous diseases and risk of cancer-associated VTE 1997 Cumulative probability of VTE (%) Cumulative probability of VTE (%) Follow-up time (days) Follow-up time (days) No prior history of phlebitis Prior history of phlebitis No prior history of varicosis Prior history of varicosis Fig. 2. Cumulative incidence of venous thromboembolism (VTE). Patients with history of superficial thrombophlebitis are compared with patients without history of superficial thrombophlebitis. Fig. 3. Cumulative incidence of venous thromboembolism (VTE). Patients with varicose veins are compared with patients without varicose veins. furthermore highly associated in the Pearson chi-squared test (P = 0.005). Additional adjustment for varicose veins resulted in a slightly reduced association between the history of superficial thrombophlebitis and the occurrence of VTE (HR = 1.82, 95% CI ; P = 0.066) (Table 3). Presence of varicose veins and association with cancer-associated VTE The condition of varicose veins was documented in 160 patients (12.6%) at study inclusion. New VTE events during the observation period were recorded in 22 (13.8%) patients with varicose veins and 76 (6.8%) patients without varicose veins. After 180 days, the cumulative probability of VTE was 5.5% for patients without and 9.4% for patients with varicose veins. After 365 days, the cumulative probabilities of VTE were 7.0% and 12.7%, respectively (Fig. 3). In univariate competing risk regression analysis, varicose veins were significantly associated with the occurrence of VTE (HR = 2.01, 95% CI ; P = 0.003). In multivariable proportional subhazards regression analysis including age, sex, obesity, newly diagnosed cancer/progression of cancer, prior history of VTE, and prior history of superficial thrombophlebitis, the association of varicose veins with the occurrence of VTE during followup remained statistically significant (HR = 2.10, 95% CI ; P = 0.003) (Table 3). Discussion Superficial thrombophlebitis and varicose veins are common conditions in cancer patients [19]. Previous studies have identified a history of superficial thrombophlebitis and the presence of varicose veins as risk factors for VTE in the general population [1,5,20]. In the present study, we have found that both a history of superficial thrombophlebitis and the presence of varicose veins also increase the VTE risk in cancer patients, about 2-fold each. Although not entirely unexpected, these associations have not previously been reported for patients with cancer. The pathophysiology of varicose veins and superficial thrombophlebitis is closely linked [21]. In view of this relationship, we performed multivariable analyses including varicose veins and history of superficial thrombophlebitis. The association of a history of superficial thrombophlebitis with the risk of VTE occurrence was mildly attenuated and was not statistically significant (Table 3). The association of varicose veins with the risk of VTE occurrence remained statistically significant. Hence, we concluded that the association of a history of superficial thrombophlebitis and VTE occurrence in cancer patients may have been dependent on the presence of varicose veins. A history of VTE is one of the strongest risk factors for new VTE events in the general population [1,7,12,22]. Cancer patients have a strongly increased risk of VTE, which might be further elevated in those with a previous history of VTE [9 11]. In our study, we were not able to confirm that a history of VTE is associated with an elevated risk of future VTE in patients diagnosed with cancer. This discrepancy between previous reports and our present study may be explained by different study designs and inclusion and exclusion criteria. For instance, patients with VTE events within 3 months prior to study inclusion were not included in our study, because we wanted to investigate the risk of new onset of cancerassociated VTE in our study without the transient effects on the hemostatic system of an acute VTE event and its anticoagulation treatment. The highest recurrence rates

6 1998 O. K onigsbr ugge et al are within the first months after a VTE event, which may explain why the impact of a history of VTE on the risk of future VTE during the course of the cancer disease was attenuated in our study. Furthermore, studies that have so far investigated the association of a history of VTE with the occurrence of further VTE events in cancer patients have not been able to distinguish whether past VTE events had occurred during or before the presence of cancer. Because extensive screening for occult malignancy in patients with unprovoked VTE is currently not recommended, it is not possible to say with absolute certainty whether a history of VTE was cancer related. One large prospective registry reported that a history of VTE increases the risk of VTE recurrence after cancer surgery but did not differentiate between VTE events occurring before or after the diagnosis of cancer [9]. Nor were VTE events specifically considered and excluded that occurred in relation to occult cancer. In our study, we attempted to exclude patients with a clear history of previous cancer-related VTE. Another study also included patients who were on therapeutic anticoagulation [8]. According to international guidelines, patients with active cancer and a history of VTE should receive long-term anticoagulation [23,24]. In our study, patients with continuous anticoagulation therapy were not eligible for study inclusion. The median time between a previous VTE event and study inclusion was 5 years (25th 75th percentile: 2 20). We recruited patients with newly diagnosed cancer and those with progressive disease after complete or partial remission. However, all patients suffered from active cancer at the time of study inclusion. Of the 540 patients who were recruited after progression of their cancer disease after remission, 37 had a history of VTE. To analyze the possibility of confounders in our study, we used a proportional subhazards model for competing risk to adjust for patients with newly diagnosed cancer and for those with progression of disease and to include the competing risk of death. Results of this statistical analysis revealed no influence on the association between a history of VTE and the risk of occurrence of VTE during the course of cancer. Limitations and strengths For organizational and logistical reasons, it was not possible to prospectively record the occurrence of new superficial thrombophlebitis during the course of cancer and to investigate the influence on the risk of VTE. The lack of active screening for VTE can be regarded as another limitation of our study. We only included symptomatic VTE events during follow-up. In other studies, the number of unsuspected VTE diagnoses during computed tomography scanning for staging was surprisingly large, and clinically asymptomatic VTE was considered to impact a cancer patient s prognosis [25,26]. Especially with regard to the association of a previous history of VTE with the occurrence of further VTE events, asymptomatic VTE events might therefore have influenced the obtained results. The low frequency of history of VTE events in our study might also have influenced the results. The number of VTE events in patient s history was further limited because patients with a recent history of VTE were not eligible, according to the exclusion criteria, thereby possibly creating a power issue. Nevertheless, the frequencies of a history of VTE, history of superficial thrombophlebitis and presence of varicose veins resemble those in similar populations of cancer patients [7,8]. We had little information concerning the severity of previous VTE events requiring anticoagulation treatment, because these patients were not eligible for the study. A strength of our study is that at study inclusion all patients were interviewed by a physician, and their history of VTE and superficial thrombophlebitis and presence of varicose veins were specifically addressed and recorded in the study database. The diagnosis of varicose veins was based on clinical evaluation. Imaging methods to establish diagnosis of presence of varicose veins were not performed. VTE as study end point was certified by an independent board of medical experts forming an adjudication committee. All suspected VTE events were presented to the committee, which independently approved potential VTE events. We performed competing risk analyses using the proportional subhazards regression according to Fine and Gray to account for the competing risk of death in our study. Conclusion A history of superficial thrombophlebitis and the presence of varicose veins led to a 2-fold increased risk of developing VTE in our cohort of a variety of cancer patients. The presence of varicose veins remained a statistically significant risk factor for VTE in cancer patients in multivariable analysis. Varicose veins may independently contribute to the overall risk of cancer-associated VTE and should be considered when assessing the risk of VTE in cancer patients. We could, however, not confirm an association between a previous history of VTE and the risk of new VTE events during the course of cancer. Future studies that clearly distinguish between the influence of a history of non cancer-associated VTE as opposed to a history of cancer-associated VTE on the risk of future VTE in cancer patients would be needed to clarify this issue. Addendum O. K onigsbr ugge acquired data, performed statistical analysis and interpreted data, drafted the manuscript, and critically revised the manuscript for important intellectual content. F. L otsch acquired data, analyzed and

7 Venous diseases and risk of cancer-associated VTE 1999 interpreted data, and critically revised the manuscript for important intellectual content. E.-M. Reitter acquired data, provided administrative and technical support, and critically revised the manuscript for important intellectual content. T. Brodowicz interpreted data and critically revised the manuscript for important intellectual content. C. Zielinski critically revised the manuscript for important intellectual content. I. Pabinger designed and conceived the study and critically revised the manuscript for important intellectual content. C. Ay designed and conceived the study, provided technical support, critically revised the manuscript for important intellectual content, and supervised the study. All authors approved the final version of the manuscript for submission. Acknowledgements We thank all persons who provided support by recruiting patients for the Vienna Cancer and Thrombosis Study. We are also thankful to the members of the adjudication committee: Renate Koppensteiner, Markus Haumer, and Andrea Willfort-Ehringer (Department of Angiology, Medical University of Vienna); Sylvia Metz-Schimmerl (Department of Diagnostic Radiology, Medical University of Vienna); and Robert Dudczak (Department of Nuclear Medicine, Medical University of Vienna). We thank Tanja Altreiter (Clinical Division of Haematology and Haemostaseology, Medical University of Vienna) for proofreading the manuscript. Further we would like to thank Florian Posch, MSc, for his valuable support in statistical analysis. Funding This work was supported by funds from the Oesterreichische Nationalbank (Anniversary Fund, project number 12739), by a grant from the Fonds der Stadt Wien f ur innovative interdisziplin are Krebsforschung, and by an unrestricted grant from Pfizer Austria. Disclosure of Conflicts of Interest The authors state that they have no conflicts of interest. References 1 Heit JA, Silverstein MD, Mohr DN, Petterson TA, O Fallon WM, Melton LJ. Risk factors for deep vein thrombosis and pulmonary embolism: a population-based case-control study. Arch Intern Med 2000; 160: Streiff MB. Anticoagulation in the management of venous thromboembolism in the cancer patient. J Thromb Thrombolysis 2011; 31: Khorana AA, Francis CW, Culakova E, Fisher RI, Kuderer NM, Lyman GH. Thromboembolism in hospitalized neutropenic cancer patients. J Clin Oncol 2006; 24: Chew HK, Wun T, Harvey D, Zhou H, White RH. Incidence of venous thromboembolism and its effect on survival among patients with common cancers. Arch Intern Med 2006; 166: Sch onauer V, Kyrle PA, Weltermann A, Minar E, Bialonczyk C, Hirschl M, Quehenberger P, Schneider B, Partsch H, Eichinger S. Superficial thrombophlebitis and risk for recurrent venous thromboembolism. J Vasc Surg 2003; 37: Galanaud JP, Genty C, Sevestre MA, Brisot D, Lausecker M, Gillet J-L, Rolland C, Righini M, Leftheriotis G, Bosson J-L, Quere I. Predictive factors for concurrent deep vein thrombosis and symptomatic venous thromboembolic recurrence in case of superficial venous thrombosis. Thromb Haemost 2010; 105: Wun T, White RH. Epidemiology of cancer-related venous thromboembolism. Best Pract Res Clin Haematol 2009; 22: Kr oger K, Weiland D, Ose C, Neumann N, Weiss S, Hirsch C, Urbanski K, Seeber S, Scheulen ME. Risk factors for venous thromboembolic events in cancer patients. Ann Oncol 2006; 17: Agnelli G, Bolis G, Capussotti L, Scarpa RM, Tonelli F, Bonizzoni E, Moia M, Parazzini F, Rossi R, Sonaglia F, Valarani B, Bianchini C, Gussoni G. A clinical outcome-based prospective study on venous thromboembolism after cancer surgery. Ann Surg 2006; 243: Srkalovic G, Cameron MG, Rybicki L, Deitcher SR, Kattke- Marchant K, Hussein MA. Monoclonal gammopathy of undetermined significance and multiple myeloma are associated with an increased incidence of venothromboembolic disease. Cancer 2004; 101: Tateo S, Mereu L, Salamano S, Klersy C, Barone M, Spyropoulos A, Piovella F. Ovarian cancer and venous thromboembolic risk. Gynecol Oncol 2005; 99: Prandoni P, Lensing AWA, Piccioli A, Bernaldi E, Simioni P, Girolami B, Marchiori A, Sabbion P, Prins MH, Noventa F, Girolami A. Recurrent venous thromboembolism and bleeding complications during anticoagulant treatment in patients with cancer and venous thrombosis. Blood 2002; 100: Ay C, Dunkler D, Simanek R, Thaler J, Koder S, Marosi C, Zielinski C, Pabinger I. Prediction of venous thromboembolism in patients with cancer by measuring thrombin generation: results from the vienna cancer and thrombosis study. J Clin Oncol 2011; 29: Ay C, Dunkler D, Marosi C, Chiriac AL, Vormittag R, Simanek R, Quehenberger P, Zielinski C, Pabinger I. Prediction of venous thromboembolism in cancer patients. Blood 2010; 116: Vormittag R, Simanek R, Ay C, Dunkler D, Quehenberger P, Marosi C, Zielinski C, Pabinger I. High factor VIII levels independently predict venous thromboembolism in cancer patients: the cancer and thrombosis study. Arterioscler Thromb Vasc Biol 2009; 29: Kurz X, Kahn SR, Abenhaim L, Clement D, Norgren L, Baccaglini U, Berard A, Cooke JP, Cornu-Thenard A, Depairon M, Dormandy JA, Durand-Zaleski I, Fowkes GR, Lamping DL, Partsch H, Scurr JH, Zuccarelli F. Chronic venous disorders of the leg: epidemiology, outcomes, diagnosis and management. Summary of an evidence-based report of the VEINES task force. Venous insufficiency epidemiologic and economic studies. Int Angiol 1999; 18: Fine JP, Gray RJ. A proportional hazards model for the subdistribution of a competing risk. J Am Stat Assoc 1999; 94: Carrier M, Khorana AA, Zwicker JI, Lyman GH, Le Gal G, Lee AY; Subcommittee on Haemostasis and Malignancy for the SSC of the ISTH. Venous thromboembolism in cancer clinical trials: recommendation for standardized reporting and analysis. J Thromb Haemost 2012; 10: van Doormaal FF, Atalay S, Brouwer HJ, van der Felde EF, B uller HR, van Weert HC. Idiopathic superficial thrombophlebitis and the incidence of cancer in primary care patients. Ann Fam Med 2010; 8:

8 2000 O. K onigsbr ugge et al 20 van Weert H, Dolan G, Wichers I, de Vries C, ter Riet G, B uller HR. Spontaneous superficial venous thrombophlebitis: does it increase risk for thromboembolism? J Fam Pract 2006; 55: Leu HJ. Phlebitides: a survey. Vasa 1994; 23: Rogers MAM, Levine DA, Blumberg N, Flanders SA, Chopra V, Langa KM. Triggers of hospitalization for venous thromboembolism. Circulation 2012; 125: Kearon C, Akl EA, Comerota AJ, Prandoni P, Bounameaux H, Goldhaber SZ, Nelson ME, Wells PS, Gould MK, Dentali F, Crowther M, Kahn SR; American College of Chest Physicians. Antithrombotic therapy for VTE Disease: antithrombotic therapy and prevention of thrombosis, 9th American college of chest physicians evidence-based clinical practice guidelines. Chest 2012; 141(suppl 2): e419s e494s. 24 Lyman GH, Khorana AA, Falanga A, Clarke-Pearson D, Flowers C, Jahanzeb M, Kakkar A, Kuderer N, Levine M, Mendelson D, Raskob G, Somerfield M, Thodiyil P, Trent D, Francis CW. American society of clinical oncology guideline: recommendations for venous thromboembolism prophylaxis and treatment in patients with cancer. J Clin Oncol 2007; 25: den Exter PL, Hooijer J, Dekkers OM, Huisman MV. Risk of recurrent venous thromboembolism and mortality in patients with cancer incidentally diagnosed with pulmonary embolism: a comparison with symptomatic patients. J Clin Oncol 2011; 29: O Connell CL, Razavi PA, Liebman HA. Symptoms adversely impact survival among patients with cancer and unsuspected pulmonary embolism. J Clin Oncol 2011; 29: