We have investigated in a prospective, randomised placebo-controlled study the effect of high-dose aprotinin on blood loss in patients admitted for major surgery (revision arthroplasty of the hip or knee, or for resection of a soft-tissue sarcoma). The mean intraoperative blood loss was reduced from 1957 ml in the control group to 736 ml in the aprotinin group (p = 0.002). The mean requirement for intraoperative homologous blood transfusion in the aprotinin group was 1.4 units (95% CI 0.2 to 2.7) and 3.1 units (95% CI 1.7 to 4.6) in the control group (p = 0.033). The mean length of hospital stay was reduced from 27.8 days in the control group to 17.6 days in the aprotinin group which was not statistically significant.

The intraoperative use of aprotinin in major orthopaedic operations significantly reduced blood loss and the required amount of packed cells. It may result in a decrease in the length of hospital stay and costs.

1 . The magnitude of the problem of congenital anomalies becomes evident when one takes into consideration the fact that they cause the death of approximately one quarter of the human race either before or shortly after birth, and handicap an appreciable proportion of the survivors throughout their lives. Further, a significant percentage of infants judged to be normal at birth are found in later life to suffer from "disguised" anomalies of the skeleton and soft tissues. Though the study of genetic factors leading to congenital defects has attracted a great deal of attention during the last few decades, the importance of environmental causes of human malformations has received relatively less emphasis. The association of congenital anomalies such as cataract and cardiac septal defects with maternal intercurrent infection of rubella during the early months of pregnancy demonstrates clearly that changes in the germplasm cannot always be invoked as the cause of developmental abnormalities. Congenital malformations that are sometimes genetically determined, such as microphthalmos, cleft palate, and certain skeletal abnormalities, can be caused in the offspring not only by maternal nutritional deficiencies and x-radiation but also, at least in some animals, such as chickens, rats and rabbits, by the introduction of certain substances like insulin into the environment of the embryo during its development.

2. Since very little is known of the detailed histology of the early human embryo, the histological examination of cases of perverted growth is mainly limited to aborted foetuses which, unfortunately, tend to present varying degrees of post-mortem degeneration before accurate histological methods can be applied. It is exactly in this field that animal experiments can offer valuable help. According to Mall and other embryologists the pathological changes that take place in human foetuses and those obtained experimentally in animals are not merely "analogous or similar but identical."

3. An attempt has been made to review, in some detail, the more important work which has been carried out on experimental teratogenesis, on the epidemiological implications of developmental arrests in humans, and on foetal abnormalities associated with maternal metabolic and hormonal disorders during pregnancy.

4. The technique employed for injection of insulin into the egg yolk has been described. Methods used for the estimation of blood sugar in chick embryos at various stages after injection of insulin and special histochemical techniques for localising polysaccharides in cartilage have been outlined.

5. A few salient experimental results have been tabulated, and some of the insulin-induced abnormalities have been illustrated.

6. The possible mechanism of action of insulin in the causation of the various developmental anomalies has been discussed. Broadly speaking, insulin seems to affect primarily the part or tissue which is in the most active stage of growth or differentiation at the time of the injection. Within the range of 0·05 to 6 units of insulin employed, the incidence, severity and distribution of the deformities appear to increase with the dose of the hormone. It has been observed that the hypoglycaemia caused by insulin injection is not counteracted till about the twelfth day of incubation, presumably because of excessive accumulation of glycogen in the yolk-sac membrane immediately after the injection, and because of lack of glycogen storage in the embryonic liver and the absence of active secretion in the endocrine glands concerned with the carbohydrate metabolism of the embryo. It has been suggested that this unchecked hypoglycaemia may deprive the mesenchyme, pre-cartilage and cartilage of glycogen and mucopolysaccharides (chondroiten-sulphuric acid complexes), depending on the time of injection and the dose of insulin, and thus not only give rise to a variety of single and multiple deformities in the cartilaginous skeleton but also interfere with the normal endochondral ossification, resulting in a generalised developmental disturbance of bone resembling osteogenesis imperfecta in the human.

7. Insulin-induced abnormalities can be prevented to a remarkable extent by injecting nicotinamide and riboflavin into eggs along with insulin.

8. The question of the practical application of the knowledge gained from experimental observations on insulin-induced developmental abnormalities in explaining the possible causation of congenital anomalies in humans by genetic and environmental teratogenic factors, has been discussed. It is suggested that the orderly progression from the mesenchymatous condensation to cartilage, and then through calcified cartilage to bone, may be disturbed by these teratogenic factors at critical phases during the development of the embryo, and a variety of single and multiple skeletal deformities may thus be induced.

9. A plea is made for routine pathological and radiological examination of aborted foetuses and stillborn infants more or less on the lines followed for experimentally induced deformities with a view to applying the knowledge gained from animal experiments to a better understanding of the etiology and pathology of human congenital anomalies.

10. As regards the possible prevention of these deformities, it is not always easy to offer sound eugenic advice in the cases of congenital malformations determined partly or completely by genetic factors, for two important reasons. First, it is often difficult to distinguish between genetically determined congenital anomalies and their phenocopies. Secondly, genetically determined developmental defects sometimes show surprisingly variable expressivity and penetrance. For the conditions in which both genetic and environmental factors are involved, the most profitable immediate line of attack would be on the environmental factors. A relatively simpler problem is presented by the malformations which are, for all practical purposes, entirely caused by environmental factors. Measures to prevent congenital anomalies caused by prenatal rubella, such as exposure of girls to the disease during childhood and protection of pregnant women during the early stages of pregnancy by immune serum, are under active consideration.

11 . Further energetic investigation of the causes of permaturity, stillbirths, monstrosities and congenital malformations is urgently needed, before embarking on a successful programme for prevention. "The day of successful prophylaxis is not yet, but it is much nearer than seemed possible a few years ago."