Major Finding: A total of 96 of 126 cases of VTE occurred more than a week after gynecologic cancer surgery.
Data Source: A nested case-control study based on a sample of 4,162 women who underwent major gynecologic cancer surgery.
Disclosures: None was reported.
SAN FRANCISCO — More than 75% of venous thromboembolisms linked to gynecologic cancer surgery occur more than a week after the procedure, based on data from 4,162 women aged 18 years and older who had undergone major gynecologic surgery between 1998 and 2008 at a single institution.
Few studies have addressed long-term postoperative risk of venous thromboembolism (VTE), and prolonged prophylaxis might be warranted for high-risk cancer patients, said Dr. Abraham Peedicayil of the Mayo Clinic in Rochester, Minn.
To determine risk factors, timing, and incidence of late VTE, Dr. Peedicayil and his colleagues conducted a nested case-control study of 126 patients who had VTE within 90 days of surgery (an incidence of 3.03%) and matched controls who did not experience VTE.
The average age of the patients was 61 years.
Overall, 30 cases of VTE occurred on days 1-7 after surgery; 50 cases, on days 8-21; and 46, on days 22-90.
Based on location of the surgery, ovarian cancer patients had nearly two-thirds (64%) of the VTEs that occurred within the first week after surgery. Patients with uterine cancers accounted for one-third (33%) of the VTEs that occurred this early, and those with cervical cancers including the cervix, vulva, and vagina made up a smaller group (3%).
But the proportion of VTEs in ovarian cancers decreased over time, to 58% during days 8-28 and 41% during days 29-90.
By contrast, the likelihood of VTE in cervical cancer patients increased over time, to 8% of all VTEs during days 8-28 and 20% during days 29-90, Dr. Peedicayil reported.
In a univariate analysis, ovarian cancer (including ovarian, tubal, and peritoneal) was an independent risk factor for VTE overall (P = .02) and on days 8-90 (P = .04).
In a multivariate analysis, independent risk factors for VTE included a previous VTE (P = .03), blood loss of 500 mL or greater (P = .044), and a hospital stay of more than 5 days (P = .044), but only the hospital stay reached statistical significance, and only for the period from days 8 to 28 after surgery (P = .002).
In all three time periods, incidence of VTEs increased with age; more than half (range, 57%-70%) occurred in women older than 60 years.
The results suggest that the majority of cases of VTE occur more than a week after surgery, beyond the time when most patients have been discharged, Dr. Peedicayil said. The findings support the use of prolonged VTE prophylaxis for high-risk patients, but more research is needed to define the highest-risk patients and to identify a more precise duration of VTE.