Summary In a consecutive series of 182 cases of hydrops fetalis 33 (17·6 per cent) were due to causes other than blood group incompatibility, the incidence of non‐immunological hydrops fetalis being 1 in 3538 deliveries. Of the 33 cases, 13 were idiopathic, 6 were associated with twin pregnancies, 7 had major malformations and the remaining 7 had a variety of associated diseases. The high incidence of polyhydramnios, pre‐eclampsia and prematurity in cases of idiopathic hydrops was confirmed, and the series included one case of eclampsia. The subsequent obstetric performance of these patients was very different to what occurs with rhesus immunization and 21 of 22 further pregnancies were uncomplicated, the other resulting in a surviving hydropic infant. There is an association (3 of 13 cases) between severe maternal anaemia and idiopathic hydrops. This association is more significant than that between hydrops fetalis and any fetal malformation other than adenomatoid hamartoma of the lung. No relationship between diabetes and of fetal hydrops was found. Urinary oestriol excretion was low in two cases of idiopathic hydrops in which this test was performed indicating that hydrops probably represents chronic feto‐placental dysfunction rather than an acute episode of circulatory failure. In non‐immunological hydrops the placental weights are of the same order as those seen in erythroblastosis. Fetal anaemia was not found in any case nor were fetal plasma protein levels consistently low. In the twin transfusion syndrome the exsanguinated twin can become hydropic and both twins may be affected. It is concluded that hydrops is unlikely to have a single aetiology and that the pathogenesis cannot be explained by such simple considerations as fetal anaemia.

No abstract

A series of 164 ovarian tumours diagnosed during pregnancy or the puerperium is presented. The commonest were the benign cystic teratoma (45 cases) and the mucinous cystadenoma (40 cases). Rapid growth of the tumour during pregnancy was observed in 5 of the 41 patients (12% ) in whom assessment was possible; in none of these was the lesion malignant. The incidence of malignancy was 2.4% (4 of 164 cases) or 10 times less than that of ovarian tumours in non-pregnant patients. This difference was accounted for by the variation in age distribution in pregnant and non-pregnant patients with ovarian tumours.There were 21 patients with tumours 15 cm. or more in diameter. In this group the incidence of malignancy was 5% and the commonest complications were torsion ( 5 patients) and infection during the puerperium ( 4 patients).

No abstract