The following has been abridged from "Twin-to-twin transfusion syndrome: New perspectives" published in The Journal of Pediatrics, November 1995. Authors: Enrico Lopriore, MD, Frank P. H. A. Vandenbussche, MD, E. Stella M. Tiersma, MD, Arnout J. de Beafort, MD, and J. Philip de Leeuw, MD, PhD.   To begin, the authors present alternative theories proposed on the causes of TTTS The presence of connected blood vessels in the shared placenta is an essential prerequisite for the development of TTTS. However, injection studies of twin placentas have shown that such anastomoses are almost universally present in monochorionic twins. Anastomoses have been documented between arteries, between veins, and between arteries and veins of each twin, but of the greatest importance for the development of TTTS are the artery-vein anastomoses that proceed through a shared placental cotyledon. The Third Circulation Described by Schatz in 1882, the 'third circulation' is the classic model of TTTS. An artery from the donor twin supplies a placental cotyledon, which is drained in turn by a vein running to the recipient. However, this classic concept of a simple shift of blood from one twin to the other has been challenged, and alternative mechanisms have been proposed. Protein Imbalance In 1963, Kloosterman found large intertwin differences in serum protein concentrations, and suggested that this difference can be attributed to a chronic loss of proteins from the donor into the circulation of the recipient. Because the placental membrane is impermeable to proteins, the hypoproteinemic donor twin, with low colloid-osmotic pressure, will then be, as it were, 'drained' by the mother, becoming dehydrated and growth retarded, whereas the hyperproteinemic recipient, with increased colloid-osmotic pressure, will be 'flooded' by fluids absorbed from the mother, which will result in increased growth, hydramnios, and, eventually, systematic edema. Increased Atriopeptin Nageotte et al. found a much higher concentration of atriopeptin in the recipient twin than in the donor. The release of this atrium derived peptide is stimulated by blood volume overload and causes increased fetal urine production, which leads to polyhydramnios. The authors suggest this may be a factor in causing the syndrome. These findings are in good agreement with the results of Wieacker et al., who, in addition, suggested that polyhydramnios is enhanced by inhibition of the release of antidiuretic hormone. Compression of Donor's Umbilical Cord Fries et al, found that in monochorionic pregnancies complicated by TTTS, a velamentous cord insertion is significantly more common in than in monochorionic gestations without the syndrome.These authors speculated that the membranously inserted cord can be easily compressed, and blood flow to one twin through the umbilical vein is reduced. Accordingly a vicious circle may be established: as less blood flows to one twin, more blood will flow to the recipient twin through the placental anastomoses. The consequent development of polyhydramnios may compress the umbilical vein even further, ultimately leading to TTS. A similar mechanism is advocated by Wax et al., who speculated that polyhydramnios may partially occlude the venous placental vessels of the donor twin. The authors suggested that large-volume amniocentesis may reduce the compressive force exerted by polyhydramnios on the placental and cord veins, thus providing not only symptomatic relief but also therapy directed at the cause. Placental Insuffiency Finally, Saunders et al. proposed that TTTS may develop as a consequence of uteroplacental insufficiency affecting the donor twin. Increased peripheral resistance in the placental circulation of the donor twin would promote the shunting of blood to the recipient through the invariably present vascular anastomoses. Verter speculated further that as a reaction to placental insufficiency and growth impairment the smaller twin may send out growth stimuli. These stimuli, however. cannot promote growth because placental insufficiency limits growth. This eventually results in increased growth of the recipient because his placental function is not impalred. The process, called the growth factor sequence, indicates why the smaller fetus may have no end-diastolic flow. The absence of end-diastolic flow velocities implies a reduction of cardiac output: thus it would hardly be possible for the donor twin to transfuse his co-twin effectively.