Ovarian cancer cells appear to be capable of both thrombin formation a
nd induction of fibrin degradation which may be essential prerequisite
s for the development of deep vein thrombosis (DVT) as well as the spr
ead of malignancy. To study further this coagulation - cancer interact
ion in 60 patients with untreated ovarian cancer of FIGO stage I-IV th
e incidence of DVT was recorded pre-operatively, post-operatively on d
ay 1, 3, 5, 7, 10, before each of six cycles of Cisplatinum/Epirubicin
/Cyclophosphamide chemotherapy, during follow-up and in the post-opera
tive period of second look surgery. In addition, blood coagulation tes
ts results were determined prospectively. Two patients were excluded f
rom these calculations due to previous DVT 5 to 6 weeks before the dia
gnosis of ovarian cancer but all patients were eligible for surgery an
d randomized to receive either daily low molecular weight heparin (LMW
H) (n = 28) or unfractionated heparin (UFH) (n = 32) for perioperative
thrombosis prophylaxis until the 7th postoperative day. According to
the FIGO stage, patients were equally distributed in the 2 heparin tre
atment groups. The predictive value of pre-operative coagulation test
results. clinical parameters, and type of heparin used were tested in
univariate and multivariate analysis for development of post-operative
DVT and overall patients survival. Impedance plethysmography for DVT
screening was used. The presence of DVT was then confirmed by phlebogr
aphy. Only D-dimer and fibrinogen levels were correlated significantly
with the FIGO stage while antithrombin, protein C, and plasminogen ac
tivator inhibitor activity were not. The incidence of DVT was 6.7% (4/
60) up to the 7th and 8.3% (5/60) between the 8th and 29th post-operat
ive day. DVT occurred in 10.6 % (5/47) during chemotherapy. pre-operat
ive coagulation test results. the type of heparin used, and clinical p
arameters were not significant risk factors for post-operative DVT dev
elopment in univariate analysis. The D-dimer and fibrinogen levels wer
e significant risk factors for reduced overall survival in univariate
analysis but only the FIGO stage was an independent predictor (in mult
ivariate analysis). after a median follow up of 26.5 months (min. 8 mo
nths, max. 41 months). 21.4% of LMWH treated and 37.5% of UFH-treated
patients died of cancer (p = 0.26). Pre-operative test results were ne
ither predictive for DVT nor the outcome of cancer but patients showed
an improved though not statistically significant overall survival aft
er LMWH treatment.