Article Text
PDF
Abstract
Introduction Venous thromboembolism prevention during the perioperative period requires comprehensive risk-level assessment. The aim of this study was to evaluate the incidence of deep vein thrombosis and to assess the cut-off levels of serum D-dimer as a screening strategy for deep vein thrombosis during the perioperative period.
Methods A total of 205 patients (ovarian cancer: 68, endometrial cancer: 76, cervical cancer: 61) who underwent gynecological surgery, including retroperitoneal lymph node dissection, were enrolled. We retrospectively analyzed the data on the cut-off value of D-dimer assessed using area under the receiver operating characteristic curve preoperatively, and 2 or 3 months, postoperatively. All patients underwent leg vein ultrasonography regardless of the serum D-dimer level. Furthermore, CT scans were performed to evaluate both disease status and venous thromboembolism, including pulmonary thromboembolism. Statistical analyzes were performed using the Mann–Whitney U-test (D-dimer values of each cancer), Chi-square test, Fisher's exact test (incidence of deep vein thrombosis), and one-way analysis of variance (patient characteristics).
Results A total of 205 patients (ovarian cancer: 68, endometrial cancer: 76, cervical cancer: 61) who underwent gynecological surgery, including retroperitoneal lymph node dissection, were included in the analysis. Deep vein thrombosis rates were significantly higher in patients with ovarian cancer (P<0.001). The postoperative D-dimer value was significantly higher than the preoperative value. Postoperative D-dimer values were also significantly higher in patients who received adjuvant chemotherapy (P=0.001). The cut-off value of D-dimer was 1.55 µg/mL preoperatively (sensitivity, 48.0%; specificity, 94.1%), and this value was higher postoperatively, at 1.95 µg/mL (sensitivity, 37.0%; specificity, 90.9%).
Conclusion Postoperative D-dimer values are higher not only after surgery but also in patients who received adjuvant chemotherapy. The cut-off value of D-dimer at 2 or 3 months postoperatively was higher than preoperative value.
- venous thromboembolism
- lymph nodes
- surgical oncology
- postoperative period
Statistics from Altmetric.com
Highlights
Deep vein thrombosis rates were significantly higher in ovarian cancer patients when compared with endometrial or cervical cancer patients.
Postoperative D-dimer values were significantly higher in patients who received adjuvant chemotherapy.
The cut-off value of D-dimer at 2 or 3 months postoperatively was higher than preoperative value.
Introduction
Venous thromboembolism often develops after surgery, and its prevention is critical. Venous thromboembolism prevention during the perioperative period requires comprehensive risk-level assessment, including patient pregnancy status, age, and surgical risk, with different management practices according to each risk level.1 Gynecological cancer is reportedly associated with a high frequency of venous thromboembolism as a preoperative complication, and it is classified as a high-risk factor for venous thromboembolism.2 3
The Japanese guidelines recommend the measurement of serum D-dimer and use of imaging tests, such as venous ultrasonography and contrast-enhanced CT, as screening tools for detecting venous thromboembolism in patients with gynecological cancer. Preoperatively, most patients undergo contrast-enhanced CT to evaluate the extent of disease, as well as the presence of venous thromboembolism: however, it is not always possible to perform venous ultrasonography for deep vein thrombosis. In clinical practice, the measurement of serum D-dimer, an end product of fibrin degradation, is often performed as a simple screening test for deep vein thrombosis. This measurement is primarily used for preoperative screening. However, the use of the cut-off value of serum D-dimer as a screening strategy for deep vein thrombosis remains unclear, particularly during the postoperative period.
In this study, we aimed to evaluate the incidence of deep vein thrombosis by the type of gynecological cancer and disease status, and assess the cut-off value of serum D-dimer as a screening strategy for deep vein thrombosis during the perioperative period in patients with gynecological cancer.
Methods
All patients with gynecological cancer underwent serum D-dimer measurement as a preoperative screening test of deep vein thrombosis at the initial visit. Moreover, all patients underwent leg vein ultrasonography regardless of the serum D-dimer level. Furthermore, contrast-enhanced CT scans were performed from the neck to the pelvic area to evaluate both the extent of disease and venous thromboembolism, including pulmonary thromboembolism (Figure 1). Patients with deep vein thrombosis were continuously administered unfractionated heparin infusion preoperatively. Anticoagulation therapy with unfractionated heparin and enoxaparin (Clexan) was administered postoperatively until discharge. Anticoagulation therapy was not administered in deep vein thrombosis-negative patients.
Flowchart of routine deep vein thrombosis screening. All patients underwent D-dimer testing and lower extremity venous ultrasonography as part of preoperative screening for deep vein thrombosis. After the surgery, the D-dimer value was again measured, and the patients underwent lower extremity venous ultrasonography if D-dimer was ≥1.0 µg/mL.
All patients who underwent gynecological surgery, including retroperitoneal lymph node dissection visited a lymphedema outpatient clinic 2 or 3 months after surgery. During this examination of lower-limb lymphedema, we measured serum D-dimer as a screening test for deep vein thrombosis, and leg vein ultrasonography was performed for patients whose serum D-dimer level was >1.0 µg/mL. As a result, patients with gynecological cancer who underwent surgery, including retroperitoneal lymphadenectomy, could examine the serum D-dimer level pre- and postoperatively. Consequently, the patients undergoing retroperitoneal lymphadenectomy were eligible in this study. The present study was approved by the Institutional Review Board of Tottori University Hospital. All patients provided written informed consent before the collection of specimens according to the institutional guidelines.
Statistical analyzes were performed using the Mann–Whitney U-test (D-dimer values of each cancer), Chi-square test, Fisher's exact test (incidence of deep vein thrombosis), and one-way analysis of variance (patient characteristics). We determined the cut-off value at the point where the sum of the sensitivity and specificity was the highest. For all tests, a P-value<0.05 was considered statistically significant.
Results
A total of 205 patients who underwent surgery for gynecological cancer, including retroperitoneal lymph node dissection, at Tottori University Hospital from 2009 to 2014 (ovarian cancer: 68 cases, endometrial cancer: 76 cases, cervical cancer: 61 cases) were included in this study (Table 1).
Patients' characteristics
Overall, the incidence of deep vein thrombosis was significantly higher in stage III–IV patients than in stage I–II patients (Table 1). Particularly in endometrial cancer, patients with stage III–IV showed significantly higher incidence of deep vein thrombosis compared with stage I–II patients (P=0.020). However, there were no differences between stage I–II and III–IV with respect to cervical or ovarian cancers.
In the preoperative period, 24/68 (35.2%) patients with ovarian cancer, 7/76 (9.2%) with endometrial, and 2/61 (3.2%) with cervical cancer had deep vein thrombosis. The incidence of deep vein thrombosis was significantly higher in patients with ovarian cancer than those with other gynecological cancer (P<0.001). In patients with ovarian cancer, the incidence of postoperative deep vein thrombosis was similar to that before surgery (P<0.001). No difference was found in the rates of pulmonary thromboembolism before and after surgery. Pathological types, Federation International Gynecologic Oncology (FIGO) stage, and residual disease did not affect the incidence of deep vein thrombosis in patients with ovarian cancer. Moreover, no significant difference in the incidence of deep vein thrombosis was found in patients with endometrial or cervical cancer.
Fourteen patients developed new deep vein thrombosis postoperatively, six (8.8%) patients with ovarian cancer, five (6.6%) patients with endometrial, and three (4.9%) patients with cervical cancer; P>0.05. Univariate analysis revealed that postoperative deep vein thrombosis onset was significantly related to age, postoperative D-dimer value, and the amount of intraoperative bleeding. The median intraoperative blood loss was 1299 (range; 75–5740) mL in patients with postoperative deep vein thrombosis, which was significantly higher than that of patients without preoperative deep vein thrombosis (755 mL (range; 100–3140); P=0.008).
The median values of D-dimer before and after surgery in all patients were 0.6 (range, 0.1–62.1) and 1.2 (range, 0.1–52.5) µg/mL, respectively: the postoperative value was significantly higher than the preoperative value (Table 2). In the endometrial and cervical cancer groups, the D-dimer values post-surgery were significantly higher than those pre-surgery. The median preoperative D-dimer values in patients with stage I–II and III–IV disease were 0.4 µg/mL (range, 0.1–62.1) and 1.1 µg/mL (range, 0.1–32), respectively: these values were significantly higher in stage III–IV cases than in stage I–II patients (P=0.02) (Figure 2A).
(A) Distribution of preoperative D-dimer values. D-dimer values were significantly higher in stage III–IV patients than in stage I–II patients (P=0.02). (B) In the adjuvant chemotherapy group, the distribution of D-dimer in all patients and in stage I–II patients. D-dimer values were significantly higher in patients who received adjuvant chemotherapy than in patients who did not (P=0.001). (C) The cut-off value of D-dimer for detecting deep vein thrombosis assessed via receiver operating characteristic curve pre-surgery was 1.55 µg/mL, and this value substantially increased to 1.95 µg/mL post-surgery.
D-dimer values and cancer type
Of 205 patients, 110 (53.6%) patients received adjuvant chemotherapy. Fewer patients with cervical cancer underwent adjuvant chemotherapy compared with patients with ovarian and endometrial cancer (P<0.001) (Table 1). The D-dimer values were significantly higher in patients who received postoperative adjuvant chemotherapy than in those who did not (P=0.001) (Figure 2B). The results of patients with stage I–II disease were similar to those of the patients overall (P=0.01).
When the cut-off value of D-dimer for detecting deep vein thrombosis was set at 1.0 µg/mL, the sensitivity and specificity were 84.8% and 69.7%, respectively. The negative predictive value for preoperative deep vein thrombosis was 96.0% and the positive predictive value was 35.0%: the postoperative values were 89.6% and 22.0% (sensitivity and specificity were 74.2% and 45.8%), respectively, which are lower than the preoperative values (Table 3). The preoperative cut-off value of D-dimer as assessed by receiver operating characteristic curve was 1.55 µg/mL, and the postoperative value was higher, which was at 1.95 µg/mL (Figure 2C).
NPV and PPV during the perioperative period
Discussion
We found that the incidence of deep vein thrombosis was significantly higher in patients with ovarian cancer than in those with other cancer types. In addition, stage III–IV patients with endometrial cancer developed deep vein thrombosis more frequently than those with stage I–II. The cut-off value of D-dimer at 2 or 3 months postoperatively was higher than that preoperatively. Although adjuvant chemotherapy did not affect the incidence of deep vein thrombosis, the D-dimer values were significantly higher in patients who received adjuvant chemotherapy after surgery than in those who did not. To the best of our knowledge, the association between deep vein thrombosis and serum D-dimer value before and after surgery in patients with gynecological cancer has not been previously reported.
Oranratanaphan et al reported that the incidence of deep vein thrombosis in gynecological cancer is highest in patients with ovarian cancer, comprising 60% of all gynecological cancer.2 Sakurai et al noted no difference in deep vein thrombosis rates among the FIGO stages in ovarian cancer.4 High rates of deep vein thrombosis and pulmonary thromboembolism have recently been reported in patients with ovarian cancer, especially in ovarian clear-cell carcinoma.5–7 Matsuura et al reported deep vein thrombosis in 27.3% of patients with ovarian clear-cell carcinoma.5 ,6 Bakhru also noted that clear-cell carcinoma and undifferentiated epithelial ovarian carcinoma are associated with a higher risk of venous thromboembolism.6 Uno et al have shown that venous thromboembolism develops prior to treatment in 45.5% of patients with ovarian clear-cell carcinoma, which is significantly higher than the rate in non-clear-cell carcinoma patients.7 Other studies have found venous thromboembolism rates of 15%–42% in ovarian clear-cell carcinoma, which are also higher than the rates in non-clear-cell carcinoma patients.8 9 However, in our study, no significant difference was found among the pathological types of ovarian cancer because the numbers with clear cell were small in terms of the venous thromboembolism rates.
Among gynecological cancers overall, few reports have been published on the distribution of deep vein thrombosis incidence by FIGO stage. In our study, over all three cancer types, the incidence of deep vein thrombosis was significantly higher in the stage III–IV cases than in the stage I–II cases. Compared with early-stage patients, D-dimer values were significantly higher in advanced-stage patients, and the incidence of deep vein thrombosis was also significantly higher in the advanced-stage patients. Therefore, in advanced-stage patients, particularly ovarian and endometrial cancer, CT scans and leg vein ultrasonography should be conducted.
Kodama et al explored the relationship between D-dimer values and venous thromboembolism after surgery in patients with gynecological cancer.10 The authors measured D-dimer values within 28 days after surgery. In contrast, we measured D-dimer values at 2 or 3 months after surgery. In the current study, all patients with pathological risk factors underwent postoperative chemotherapy within at least 2 months after surgery. Consequently, it is important to consider whether postoperative adjuvant chemotherapy might affect D-dimer values or not.
Several studies have reported an association between chemotherapy and thrombosis. von Tempelhoff et al noted that although monitoring of patients using sophisticated coagulation tests did not identify patient risk of deep vein thrombosis during chemotherapy, evaluation of the coagulation status before initiating chemotherapy is recommended.11 Samare F et al reported that deep vein thrombosis incidence might be higher in patients who underwent chemotherapy than in patients who underwent surgery (P=0.13).12 Further, regarding the effects of chemotherapy on the incidence of deep vein thrombosis, Kirwan et al examined that chemotherapy-induced apoptosis of endothelial cell may enhance hypercoagulability and initiate venous thromboembolism.13 However, they have concluded that chemotherapy-induced venous thromboembolism is not primarily due to endothelial cell activation. It may be difficult to mention the effects of chemotherapy on D-dimer directly. In our study, although the D-dimer values were significantly higher in the chemotherapy group than in the non-chemotherapy group, differences were not observed between these groups regarding the incidence of deep vein thrombosis.
Various reports have been published on the cut-off value of preoperative D-dimer. Kawaguchi et al proposed a cut-off value for D-dimer before treatment of 1.5 µg/mL in ovarian cancer (87 cases): when the D-dimer value is higher than this cut-off value, patients should undergo a lower extremity venous ultrasonography.14 Although not studied in the cancer types in the present study, the cut-off value of D-dimer for preoperative deep vein thrombosis was 1.55 µg/mL, which is similar to the cut-off value proposed by Kawaguchi et al. Interestingly, using receiver operating characteristic curve calculation, this value increased to 1.95 µg/mL postoperatively.
Our study has some limitations. First, the subjects were limited to cases in whom pelvic lymph node dissection was performed: thus, the incidence of deep vein thrombosis onset and the D-dimer values are not representative of all patients with gynecological cancer. Second, we did not evaluate fibrin degradation product and other coagulation parameters or liver and renal functions. Third, D-dimer was measured 2 or 3 months post-surgery but not at every month post-surgery: measuring D-dimer immediately after and every month after surgery could increase the accuracy of assessing deep vein thrombosis onset and cause retrieval. Further, the use of anticoagulant has not been considered in this study. Detailed data on the onset of deep vein thrombosis can be obtained by examining whether or not therapeutic and prophylactic anticoagulant therapy is available.
In summary, our study showed that D-dimer values are significantly higher post-surgery than pre-surgery. D-dimer values were significantly higher in the adjuvant chemotherapy group, suggesting that postoperative D-dimer values are affected not only by surgery but also by chemotherapy. Further, the cut-off value of D-dimer as assessed by receiver operating characteristic curve pre-surgery was 1.55 µg/mL, and this value increased substantially to 1.95 µg/mL post-surgery.
Acknowledgments
Enago provided language help, writing assistance, and proofreading services for this article. We would like to express our sincere gratitude to them.
References
Footnotes
Contributors HK and MS devised the project, the main conceptual ideas, and proof outline. HK and MS worked out almost all of the technical details and performed numerical calculations for the suggested experiment. HK wrote the manuscript. All authors provided critical feedback and helped shape the research, analysis, and manuscript.
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial, or not-for-profit sectors.
Competing interests None declared.
Patient consent for publication Not required.
Provenance and peer review Not commissioned; externally peer reviewed.
Data availability statement Data are available in a public, open access repository. Data are available in a public, open access repository.