Recently, observational studies have suggested an increased risk of no
nfatal venous thromboembolic complications in women using oral contrac
eptives (OCs) containing the third-generation progestins, gestodene an
d desogestrel. Because of the observational, rather than randomized, n
ature of these trials, the clinical relevance of these findings is dif
ficult to interpret. Each study included one or more potential sources
of bias. In particular, cases came almost exclusively from hospitaliz
ed patients with nonfatal venous thromboembolism, which represents onl
y a minority of patients diagnosed with this condition according to cu
rrent clinical practice. In the absence of a sound biologic rationale
to explain the increased risk with third-generation OCs, and consideri
ng the potential sources of bias within the current studies, an altern
ative view argues against causality. Oral contraceptives remain safe a
nd effective. Clearly, additional research is needed to determine the
relationship between thromboembolic disease and the use of third-gener
ation OCs. In the interim, women should be informed thoroughly with ob
jective data on all risks associated with the use of OCs. (C) 1997 by
The American College of Obstetricians and Gynecologists.