Malignant sacrococcygeal teratoma--a problem in diagnosis.

and Kendall, 1966) as well as a rise and alteration in plasma lipids. In addition, a decrease in fibrinolysis may be present (Cotton, 1967). Hypovolaemia is a usual finding in the nephrotic syndrome, and may be made worse by diuretic therapy. The increase in haematocrit, the vaso-constriction, and sluggish circulation tend to increase the likelihood of intravascular coagulation. Coagulation is also promoted by corticosteroid therapy in both nephrotic and non-nephrotic patients (Ozsoylu, Strauss, and Diamond, 1962). Some coagulation factors are increased, particularly factor VIII. It is not surprising that throm-botic incidents in nephrotic patients were noted, for example by Addis (1949). A curious anomaly has been the relatively frequent observation of pulmonary artery thrombosis in children (Levin, Zamit, and Schmaman, 1967). Finally there is little doubt that there is a risk of trauma to the femoral artery during attempts at femoral vein puncture. This would be particularly likely to occur in a shocked, nephrotic patient where venoconstriction and oedema would make it difficult to obtain a sample. Goldbloom, Hillman, and Santulli (1967) reported three nephrotic children receiving corticosteroids in whom femoral vein puncture was followed by thrombosis of the femoral artery and amputation, with death in one case. Femoral artery thrombosis has also been reported after femoral vein puncture in non-nephrotic infants (Nabseth and Jones, 1963). It is difficult to assess the amount of risk; thousands of femoral vein punctures are performed in children every year without incident (McKay, 1966), and probably some dozens of these are on nephrotic children. However, spontaneous arterial throm-botic episodes in children who do not suffer from embolism are extremely rare. It seems reasonable to us that femoral vein puncture should be avoided in nephrotic children unless the sample is vital and cannot be obtained elsewhere. In view of the good results obtained with cyclo-phosphamide in children with the 'minimal change' lesion and the nephrotic syndrome (Moncrieff et al., 1969), it seems reasonable to add thrombotic episodes to the indications for transferring patients from corticosteroid therapy to this drug. Our patient had, in any case, severe steroid toxicity and increasing resistance to treatment. Summary A 2*-year-old girl with the nephrotic syndrome is described. During a relapse she developed femoral artery thrombosis leading to a through knee amputation. At the time of the incident the patient was hypotensive with a raised blood urea, was receiving corticosteroids, and had bilateral femoral vein punctures to obtain blood for …