Objective: To determine the prevalence and clinical significance of abnormalities of preoperative coagulation tests in gynecologic oncology patients.
Methods: Three hundred fifty-one patients presenting for inpatient surgical procedures on the gynecologic oncology service at Duke University Medical Center from January 1, 1990 to December 31, 1990, underwent preoperative coagulation testing. Twenty-nine patients had only prothrombin time (PT) and partial thromboplastin time (PTT) measured; the remaining 322 had preoperative measurement of PT, PTT, fibrinogen, and fragment D-dimer. Outcomes assessed were perioperative hemorrhage resulting in death or reoperation, postoperative hematomas, and need for intraoperative and postoperative transfusion.
Results: Twelve of 351 patients (3.4%) had abnormally elevated PT or PTT; six of these were attributable to risk factors unrelated to malignancy. One hundred fifty-six of 322 subjects (48.4%) had abnormal levels of fibrinogen, mostly elevations above 360 mg/dL, and 88 of 322 subjects (27.3%) had positive tests for D-dimer. Fifty-seven (17.7%) had both elevated fibrinogen and positive D-dimer. One hundred eighty-eight of 322 subjects had at least one abnormal test result. There were no perioperative deaths or reexplorations because of hemorrhage. There was one postoperative hematoma. The combination of an elevated fibrinogen and a positive D-dimer test was a significant predictor of perioperative transfusion in a logistic regression model incorporating stage, preoperative hematocrit, and age (odds ratio 1.96, 95% confidence interval 1.03-3.76). However, the attributable risk associated with this abnormality was only 7.7% in patients at highest risk of transfusion.
Conclusion: Although abnormalities in coagulation are common in patients undergoing surgery for gynecologic malignancy, preoperative testing for occult coagulopathy provides little clinically useful information.