Cancer is a well-known risk factor for the development of venous thromboembolism (VTE). Malignancy results in the activation of the coagulation system and a prothrombotic state.1 Therapeutic interventions associated with malignancy, such as surgery and chemotherapy, increase the risk for development of VTE.2,3 The VTE in cancer patients necessitates long-term anticoagulation (often times at a significant cost), a 12% annual risk for bleeding complications and up to a 21% risk of recurrent events each year, and potentially delays surgical interventions or chemotherapy.4 It is the second leading cause of death in patients with cancer. Cancer patients with VTE have a 2-fold greater risk for death compared with cancer patients without VTE.5 Patients with cancer who are currently undergoing treatment are at the highest risk for development of VTE. Whereas cancer patients have a greater than 4-fold increase in the risk for VTE compared with the general population, patients actively on chemotherapy have a greater than 6-fold increased risk.6
The VTE is increasingly being diagnosed incidentally on imaging studies, such as CT scans, done for routine cancer surveillance. These “incidental” VTEs, although potentially not having the symptomatic impact of other events, seem to be associated with poorer survival, as evidenced by a recently published report that demonstrated a hazards ratio for death of 1.51 (95% confidence interval, 1.01-2.27; P = 0.048).7 A study in patients with lung cancer demonstrated a worse overall survival for both incidental VTE (HR, 2.4; P = 0.01) and VTE diagnosed due to symptoms (HR, 2.7; P < 0.002).8
Gynecologic cancers have been deemed high risk for the development of VTE by a recently validated predictive model for the development of VTE in cancer patients. Malignancy results in a predisposition to the development of VTE via numerous factors, including release of procoagulant agents by tumor cells, vascular stasis, and vascular injury. Ovarian cancer typically presents in the advanced stage with widely metastatic disease. The usual treatment paradigm for advanced ovarian cancer is a combination of aggressive cytoreductive surgery with the goal of maximal tumor debulking and adjuvant chemotherapy, usually involving a combination of platinum and taxane agents.9 In 2005, a retrospective review of 253 patients with ovarian cancer treated with surgery and chemotherapy in Italy between 1990 and 2001 revealed a 16.6% rate of VTE. Multivariate analysis demonstrated that residual disease, increasing age, and BMI were independent risk factors for the development of VTE.10 In 2007, investigators from UC Davis reported that, among 13,031 cases of ovarian cancer diagnosed in California between 1993 and 1997, there was a 5.2% rate of VTE within 24 months of diagnosis. Development of VTE was a predictor of reduced survival.11
Thromboprophylaxis is recommended for all inpatients with cancer based on recommendations from the American Society of Clinical Oncology and National Comprehensive Cancer Network guidelines. These recommendations are based on large trials of patients requiring hospitalization, including a small percentage of patients with malignancy. As cancer patients are increasingly being treated on an outpatient basis, the question arises: “Should these patients undergo active prophylaxis during treatment, even when they are not in the hospital?” Two large, randomized trials have been performed examining the use of prophylactic low-molecular-weight heparin used on an outpatient basis in cancer patients undergoing chemotherapy. The PROTECHT study examined 1150 patients with locally advanced or metastatic lung, gastrointestinal, pancreatic, breast, ovarian, or head/neck cancers. There was a decrease from 3.9% to 2.0% in the rate of VTE in patients who underwent daily prophylaxis (1-sided 95% CI, 0.303%; P = 0.02).12 There were 143 patients with ovarian cancer included in the PROTECHT study; however, details regarding stage, surgery, chemotherapy, and VTE rate were not reported in this specific disease. The SAVE-ONCO trial was published in 2012. This was a prospective, double-blind trial of more than 3200 patients with locally advanced or metastatic lung, pancreatic, gastric, colorectal, bladder, or ovarian cancer who were randomized to receive daily prophylaxis with subcutaneous semuloparin or placebo. There was a decrease in the rate of VTE from 3.4% to 1.2% (P < 0.0001) in those patients who received prophylaxis.13 Of the 191 ovarian cancer patients who were treated with semuloparin, one developed VTE. None of the 188 ovarian cancer patients treated with placebo had a thromboembolic event.
Outpatient prophylaxis in patients with cancer undergoing active treatment has not been widely adopted, and this is likely due to a lack of understanding of the scope of the problem, especially in different disease states.14 A recently published cohort study of greater than 17,000 cancer patients undergoing outpatient treatment found a 12.6% rate of VTE for 12 months after the initiation of chemotherapy compared with 1.4% in the control group (P < 0.0001).15 Rates of VTE ranging between 5% and 7% have been found in acutely ill hospitalized patients or postoperative patients, and these rates have been deemed high enough to warrant prophylaxis in these patients.16,17
Fotopoulou et al18 attempted to elucidate the patterns of VTE in advanced ovarian cancer patients treated with first-line platinum and taxane-based chemotherapy in a clinical trial setting. A total of 2743 patients were included in the analysis, and 76 (2.8%) experienced a clinically apparent VTE during the study period. On multivariate analysis, increasing age and BMI of greater than 30 kg/m2 were found to be significant predictors of VTE (HR, 3.2; 95% CI, 2.0–5.2). The patients in this cohort did not receive extended-duration prophylaxis after surgery. Patients who developed symptomatic pulmonary emboli in this study did have overall poorer survival outcomes with 1-year mortality from diagnosis of 21% compared with 9%, but other types of VTE did not seem to confer a poorer prognosis.
Mereu et al19 reported on their experience with ovarian cancer patients undergoing frontline chemotherapy and the associated risk for VTE. They published a retrospective review in 2009 of 203 patients treated from 1990 to 2004. The risk for symptomatic VTE was 7.8% at 6 months. On multivariate analysis, BMI, histology, single-agent chemotherapy, and FIGO stage were all predictive of VTE. A report from Levitan et al20 on hospitalized Medicare cancer patients who required hospitalization found that the highest rates of VTE were in patients with ovarian cancer.
We sought to determine the prevalence of VTE in a population of advanced ovarian cancer patients treated at an academic medical center. The combination of metastatic spread at presentation, major abdominal surgery with associated morbidity, and rapid commencement of adjuvant chemotherapy (given on an outpatient basis) increases the potential for VTE in these patients. Minimal data exist regarding the rates of VTE in ovarian cancer patients during adjuvant chemotherapy when the patients have been treated with appropriate, extended-duration postoperative thromboprophylaxis.
A retrospective chart review was performed from an IRB-approved database of advanced epithelial ovarian cancer patients who underwent surgery and chemotherapy at the Northwestern Memorial Hospital from January 2008 through December 2011. Advanced-stage disease was defined as FIGO stage IIC, III, or IV. Patients were excluded from analysis if they had a previous VTE (diagnosed before up-front surgical cytoreduction or commencement of neoadjuvant chemotherapy) or VTE diagnosed in the immediate postoperative period (defined as within 28 days of surgery or before initiation of adjuvant chemotherapy). All types of epithelial ovarian cancer patients were included except for those with clear cell histology because of the previously described association with VTE in this disease subtype.21 Patients were also excluded from analysis if they were on anticoagulation for a different indication. Starting in 2008, it became part of the standard postoperative protocol for patients who underwent laparotomy for gynecologic malignancy at the Northwestern Memorial Hospital to complete 28 days of prophylactic low-molecular-weight heparin. Only those patients who remained compliant with the 28 days of prophylactic anticoagulation were included in this study.
Patient demographics, operative results and procedures performed, perioperative complications, and oncologic outcomes were extracted from the electronic medical record. Demographics included age, BMI, and medical comorbidities. Operative results included surgical procedures, residual disease after surgery, estimated blood loss, and length of hospital stay. Complications that occurred within 30 days of surgery were analyzed. Included in the analysis were VTE events (defined as deep vein thrombus [DVT] or pulmonary embolus [PE]), infection (wound infection, vaginal cuff cellulitis/abscess, pelvic abscess, pneumonia, sepsis), vascular injuries, ileus/small bowel obstruction, and renal issues (ureteral injury, acute tubular necrosis, urinary tract infection). Oncologic outcomes analyzed included response type and rate, progression-free survival, overall survival, type of adjuvant treatment, and final FIGO stage. Progression-free survival was calculated from the date of surgery until the date of the first disease progression or death and date of the last follow-up if the patient was still alive. Overall survival was calculated from the date of surgery until death or last follow-up. The presence of intraperitoneal or tunneled venous catheters was noted. The VTE was generally diagnosed through 1 of 2 different algorithms. When there was a clinical suspicion for an event based on patient symptoms or physical examination findings, a DVT was diagnosed using a duplex compression ultrasound, and a PE was diagnosed using a spiral CT scan. In addition, VTEs were incidentally diagnosed during CT scans performed at the conclusion of adjuvant chemotherapy to document disease response and status. Patients with superficial thrombophlebitis were not included.
Statistical analysis was performed using SPSS software (Chicago, IL). There were 17 different variables analyzed. Fisher exact test was used to analyze the dichotomous variables. Wilcoxon rank-sum test was used for the continuous variables. Statistically significant indices on univariate analysis were included in a Cox proportional hazards multivariate regression analysis to predict VTE risk. Kaplan-Meier survival curves were used to analyze progression-free and overall survival. Findings were considered statistically significant if there was a P ≤ 0.05.
During the study period, from January 2008 through December 2011, 128 patients were identified, who met all of the inclusion criteria. Overall cohort patient data are summarized in Table 1. The mean age for the whole cohort was 60.6 (SD, 11) years, and the mean BMI was 26.7 (SD, 5) kg/m2. High-grade papillary serous histology was present in 123 patients (96%), whereas the remaining 5 patients (4%) had high-grade endometrioid histology. Seventy-nine of the patients (62%) had stage IIIC disease, 22 (17%) had stage IV disease, and the remaining 17 patients had IIC, IIIA, or IIIB disease. The rate of optimal cytoduction, defined as residual disease of less than 1 cm, was 88% in the study population. The mean length of postoperative stay for all patients was 5.6 (SD, 3) days. All patients were treated with platinum and taxane-based chemotherapy for a median of 6 (range, 4-9) cycles. Eighty patients (63%) were treated with intravenous platinum/taxane chemotherapy given every 21 days. Four patients (3%) received intravenous platinum delivered every 21 days and intravenous paclitaxel given weekly in a dose-dense fashion. Forty-four patients (34%) were treated with a combination of intraperitoneal and intravenous platinum/taxane chemotherapy. Fifteen patients (12%) were treated with neoadjuvant chemotherapy before surgical intervention. Forty-seven patients (37%) had a tunneled venous catheter in place that was used for chemotherapy administration and serum laboratory draws, and 60 patients (47%) had an intraperitoneal port placed at the time of surgery or in the immediate postoperative period by interventional radiology. Ninety-nine of the patients (77%) had a complete response to frontline therapy based on a combination of physical examination findings, serum CA-125 level, and CT scans performed at the conclusion of treatment.
Sixteen patients had a reported VTE during the time they were on frontline chemotherapy (12.5%). Nine patients (7%) had a PE, and 8 (6.3%) had a DVT. One patient was diagnosed with, simultaneously, a PE and a lower-extremity DVT. Twelve (75%) of the 16 VTEs were clinically apparent and were diagnosed based on physical examination findings or patient symptoms that led to diagnostic imaging. Four of the events (25%) were diagnosed on routine imaging done at the conclusion of treatment to assess disease status and response to therapy. There was no significant difference in the age or BMI of the patients in the group that developed a VTE compared with the group that did not (62 vs 60 years, P = 0.40) (26.5 vs 28 kg/m2, P = 0.23). There was also no significant difference in the rates of suboptimal cytoreductive surgery, presence of a tunneled venous catheter or intraperitoneal port, use of intraperitoneal chemotherapy, rates of complications associated with chemotherapy, or use of bevacizumab or weekly (dose-dense) paclitaxel between the 2 patient cohorts (Table 2). Patients who developed VTE during chemotherapy did have a significantly longer postoperative stay (8.5 days) compared with patients who did not develop VTE (5.25 days, P = 0.001). There was a significantly higher rate of perioperative complications in the VTE group (56%) compared with the control group (40%, P = 0.04). The patients who developed VTE had a significantly lower complete response rate to chemotherapy (63%) compared with the group with no VTE (80%, P = 0.02). There was a trend toward a higher preoperative CA-125 (2624 vs 1298, P = 0.06) and a higher rate of bowel resection and/or splenectomy during surgery (38% vs 16%, P = 0.07) in the group of patients who developed VTE. There was also a trend toward poorer median progression-free survival in the cohort that developed VTE (13 months) compared with the control group (24 months, P = 0.19). A similar trend was noted for 3-year overall survival. Patients who developed VTE had a 51% 3-year OS compared with 73% with no VTE (P = 0.14). On multivariate analysis, postoperative length of stay and response to chemotherapy remained statistically significant predictors of development of VTE in this patient population.
The results of our investigation into the rates of VTE in advanced, epithelial ovarian cancer patients undergoing frontline adjuvant chemotherapy showed a 12.5% rate of VTE. Previous studies of thromboembolic events in ovarian cancer patients undergoing chemotherapy have demonstrated rates of VTE ranging from 5% to 16%, and our results are consistent with these previous reports. Given the potential for acute and long-term morbidity (and possible mortality), especially in this patient population with a high rate of recurrence and continued need for treatment, this raises the question of need for extended prophylaxis for the entire duration of treatment. Postoperative rates of DVT in gynecologic cancer patients have been reported as high as 38%, whereas rates of pulmonary embolism have been reported as high as 6.8% in ovarian cancer patients. These results have led to a dramatic shift in treatment paradigm where postoperative, inpatient prophylaxis is increasingly common and is becoming part of the standard of care in this patient population. One report found that 76% of postoperative VTE in cancer patients were found greater than 1 week after surgery. Results such as these have spurred many to use extended-duration (28-day) postoperative prophylaxis.22 On the basis of these results, we began using extended-duration (28-day) low-molecular-weight heparin prophylaxis in all of our major cancer surgeries on the gynecologic oncology service in 2008. Previous reports of VTE in ovarian cancer patients undergoing chemotherapy have not had rigid guidelines for extended-duration postoperative prophylaxis, and this may cloud their results by falsely increasing the rate of VTE thought to be due to chemotherapy when, in fact, these were postoperative events.
Increasing age and BMI, often thought to be predictors of VTE risk, were not found to play a role in our study. Previous reports have found that bulky residual disease after cytoreductive surgery was a major risk factor for the development of VTE, but patients who underwent suboptimal cytoreduction were not found to have significantly higher rates of VTE. Intraperitoneal ports or tunneled venous catheters, often thought to increase an individual’s risk for VTE, were not associated with a higher rate of blood clots. The use of bevacizumab or weekly paclitaxel, both of which act via inhibition of angiogenesis and may increase the risk for VTE, were not found to do so in this patient cohort. Neither treatment with intraperitoneal chemotherapy nor complications associated with chemotherapy requiring hospitalization were associated with increased risk for VTE.
Factors that were associated with an increased rate of VTE seemed to be related to either the initial surgical effort, the inherent biology of the disease, or a combination of these two. Development of VTE was tied to a longer postoperative stay, both on univariate and multivariate analysis. Although this, inherently, makes sense, the use of postoperative VTE prophylaxis should have obviated this difference. Higher rates of perioperative complications were found in the group of patients who developed VTE on univariate analysis. Again, it seems as though the use of extended-duration prophylaxis did not decrease the rates of VTE in this population. Perhaps, an even higher rate would have developed in the absence of prophylaxis, but this conclusion is not verifiable. Multivariate analysis also revealed response to chemotherapy to be related to risk for VTE. Patients who had a complete response to treatment had a significantly lower rate of VTE than patients who did not. We hypothesize that perhaps the long-term, continued presence of residual cancer results in a continued activation of malignancy-induced prothrombotic cascades that place these patients at higher risk for VTE. A trend toward higher preoperative CA-125 and the need for bowel resection and/or splenectomy during surgery suggest that, perhaps, ovarian cancer that is more biologically “aggressive” may predispose patients to developing VTE. Both of these factors, however, may simply be indicative of a more radical surgery due to the bulk of disease, and this could be directly linked to a longer postoperative stay.
Although most events in our cohort (75%) were diagnosed based on clinical findings such as lower extremity pain or swelling, chest pain, shortness of breath, or dyspnea on exertion, a substantial portion (25%) were found on routine imaging done for cancer surveillance. This result highlights the potential for VTE to be present even in the absence of demonstrable symptoms and the need for continued vigilance.
This study is limited by its relatively smaller numbers and retrospective nature. The smaller number of patients also detracts from the quality of the multivariate model. The strengths of this study include the universal extended-duration postoperative prophylaxis that patients underwent to be eligible. In addition, patients were generally treated on similar protocols by a group of gynecologic oncologists at 1 institution with similar philosophies and treatment paradigms for advanced ovarian cancer. In our view, a 12.5% rate of VTE in advanced ovarian cancer patients undergoing frontline adjuvant chemotherapy warrants consideration of prophylaxis in these patients during their entire course of treatment, especially patients with bulky disease who undergo aggressive surgical cytoreduction with a prolonged hospital course followed by a nonoptimal response to adjuvant chemotherapy. No survival benefit has been demonstrated in a cancer patient population when long-term VTE prophylaxis is performed. Further investigation with prospective, randomized studies is required for a definitive answer to this perplexing question.
1. Khorana A, Francis CW, Culakova E, et al. Risk factors for chemotherapy-associated venous thromboembolism in a prospective observational study. Cancer
. 2005; 104: 2822–2829.
2. Khorana A Cancer-associated thrombosis: updates and controversies. Thrombosis in Challenging Populations. Hematology
. 2012; 1: 626–630.
3. Lip G, Chin BS, Blann AD, et al. Cancer and the prothrombotic state. Lancet Oncol
. 2002; 3: 27–34.
4. Prandoni P, Lensing AW, Piccioli A, et al. Recurrent venous thromboembolism and bleeding complications during anticoagulation treatment in patients with cancer and venous thrombosis. Blood
. 2002; 100: 3484–3488.
5. Sorensen H, Mellemkjaer L, Olsen JH, et al. Prognosis of cancers associated with venous thromboembolism. N Engl J Med
. 2000; 343: 1846–1850.
6. Chew H, Wun T, Harvey D, et al. Incidence of venous thromboembolism and its effect on survival among patients with common cancers. Arch Intern Med
. 2006; 166: 458–464.
7. O’Connell C, Razavi P, Ghalichi M, et al. Unsuspected pulmonary emboli adversely impact survival in patients with cancer undergoing routine staging multi-row detector computed tomography scanning. J Thromb Haemost
. 2011; 9: 305–311.
8. Connolly G, Dalal M, Lin J, et al. Incidental and symptomatic venous thromboembolic events are associated with increased mortality in lung cancer patients. J Thromb Haemost
. 2011; 52: 713–718 .
9. Bristow R, Tomacruz RS, Armstrong DK, et al. Survival effect of maximal cytoreductive surgery for advanced ovarian carcinoma during the platinum era: a meta-analysis. J Clin Oncol
. 2002; 20: 1248–1259.
10. Tateo S, Mereu L, Salamano S, et al. Ovarian cancer and venous thromboembolic risk. Gynecol Oncol
. 2005; 99: 119–125.
11. Rodriguez A, Wun T, Chew H, et al. Venous thromboembolism in ovarian cancer. Gynecol Oncol
. 2007; 105: 784–790.
12. Agnelli G, Gussoni G, Bianchini C, et al. Nadroparin for the prevention of thromboembolic events in ambulatory patients with metastatic or locally advanced solid cancer receiving chemotherapy: a randomised placebo-controlled, double-blind study. Lancet Oncol
. 2009; 10: 943–949.
13. Agnelli G, George DJ, Kakkar AK, et al. Semuloparin for thromboprophylaxis in patients receiving chemotherapy for cancer. N Engl J Med
. 2012; 366: 601–609.
14. Lyman G Venous thromboembolism in the patient with cancer: focus on burden of disease and benefits of thromboprophylaxis. Cancer
. 2011; 117: 1334–1349.
15. Khorana A, Dalal M, Lin J, et al. Incidence and predictors of venous thromboembolism among ambulatory high-risk cancer patients undergoing chemotherapy in the United States. Cancer
. 2012; 119: 1–8.
16. Samama M, Cohen AT, Darmon JY, et al. A comparison of enoxaparin with placebo for the prevention of venous thromboembolism in acutely ill medical patients: Prophylaxis in Medical Patients with Enoxaparin Study Group. N Engl J Med
. 1999; 341: 793–800.
17. Leizorovicz A, Cohen AT, Turpie AG, et al. Randomized, placebo-controlled trial of dalteparin for the prevention of venous thromboembolism in acutely ill medical patients. Circulation
. 2004; 110: 874–879.
18. Fotopoulou C, duBois A, Karavas AN, et al. Incidence of venous thromboembolism in patients with ovarian cancer undergoing platinum/paclitaxel-containing first-line chemotherapy: an exploratory analysis by the Arbeitsgemeinschaft Gynaekologische Onkologie Ovarian Cancer Study Group. J Clin Oncol
. 2008; 26: 2683–2689.
19. Mereu L, Tateo S, Klersy C, et al. Stratification of venous thromboembolism risk in ovarian cancer patients during chemotherapy. Int J Gynecol Cancer
. 2009; 19: 79–83.
20. Levitan N, Dowlati A, Remick SC, et al. Rates of initial and recurrent thromboembolic disease among patients with malignancy versus those without malignancy. Risk analysis using Medicare claims data. Medicine
. 1999; 78: 285–291.
21. Duska L, Garrett L, Henretta M, et al. When “never-events” occur despite adherence to clinical guidelines: the case of venous thromboembolism in clear cell cancer of the ovary compared with other epithelial ovarian cancer subtypes. Gynecol Oncol
. 2010; 116: 374–377.
22. Clarke-Pearson D, Abaid LN Prevention of venous thromboembolic events after gynecologic surgery. Obstet Gynecol
. 2012; 119: 155–167.