To assess the potential value of intrauterine diversion in cases of fetal urinary tract malformation, we re-evaluated the 19 cases of such lesions which were part of a consecutive series of 94 cases of fetal malformations that were diagnosed in our department between 1980 and 1985. We chose a policy of non-interference for two reasons: Firstly the relative paucity of available data compared with what one would normally expect to obtain when planning a treatment of the newborn, and secondly the expectation that the benefit in terms of recovery of function would be minimal, because a lesion observable on even early echography is more than trivial, and no longer "early" in an ontogenic sense. The empirical criteria which indicate that intrauterine intervention is either futile or superfluous, noted in the literature, were applied in a retrospective study of our cases and evaluated in relation to the data of pregnancy duration, clinical outcome, and pathological findings in the urinary tract and other systems.(ABSTRACT TRUNCATED AT 250 WORDS)