Amniotic Septostomy for the Treatment of Twin Oligohydramnios Polyhydramnios Sequence - George Saade et al ABSTRACT December 1997 OBJECTIVE: To report our experience with intentional puncture of the intervening membrane ("septostomy") for the treatment of the twin oligohydramnios-polyhydramnios sequence (TOPS) METHODS: 12 patients were diagnosed with TOPS based on ultrasonographic findings. A 20 to 22-gauge spinal needle was used to puncture the membrane between the twins without any attempt at amnioreduction in 9 patients, while the procedure was combined with amnioreductions in 3 patients RESULTS: Gestational age was 23.1 +/- 3.3 weeks at the time of septostomy and 31.1 +/- 4.4 weeks at delivery. Rapid accumulation of fluid around the "stuck" fetus occurred in all cases following a single procedure. Three of the 24 fetuses died in utero and one died on the fifth day of life, for a combined survival of 83.3%. In the survivors, the septostomy to delivery interval ranged between 0.6 and 13 weeks (mean + SD: 8.3 +/- 4.8 weeks) CONCLUSION: Amniotic septostomy is a promising new method for the management of TOPS and is associated with survival rates that are better than, or comparable to, more invasive modalities. A multicenter trial comparing septostomy to other modalities is warranted. Extract from the report:  The finding of markedly increased amniotic fluid in the gestational sac of one twin and virtual absence of fluid in the other's is an ominous complication of monochorionic-diamniotic twin pregnancies. Because of the belief that this derangement is caused by placental vascular communications (chorioangiopagus) the condition has been referred to as twin-twin transfusion syndrome. Actual shunting of blood between the twins, however, has not been documented in all cases which have fulfilled the ultrasonographic criteria. For this reason, some investigators prefer to use descriptive terms such as the "stuck twin syndrome" or "twin oligohydramios-polyhydramnios sequence" (TOPS) to refer to this condition. Due to differing diagnostic criteria and ascertainment bias, the estimated incidence of this syndrome in twin gestations has ranged widely between 1 - 26%. The data on perinatal mortality have also been affected by diagnostic inconsistencies and the general improvements in perinatal survival through the years. In the untreated condition, the mortality rate approaches 100%, and prognosis is especially poor when diagnosed in the second trimester. Given this high perinatal mortality, several aggressive treatment modalities have been attempted including selective feticide, hysterotomy for umbilical cord ligation or extirpation of one twin, bloodletting from a placental vessel, and maternal digoxin therapy. More recently, fetoscopic-guided laser ablation of the chorioangiopagus placental vessels has been reported to improve survival. Serial decompression amniocentesis, however, remains the most widely used therapy. In a recent review of the literature, Moise reported an overall survival rate of 49% with serial amnioreduction. In most cases, the polyhydramnios recurs after a few days to weeks and repeated decompression amniocenteses are required. However, several reported series include a number of cases in which the stuck twin syndrome resolved after a single amniocentesis or suddenly following an amniocentesis that was the last in a series, with no need for additional therapy. Equalization of amniotic fluid volumes following a single amniocentesis was first reported by Wax et al. in a patient diagnosed at 27 4/7 weeks' gestation. An ultrasound performed five days after removal of 1,300 ml of amniotic fluid revealed polyhydramnios in both sacs. The ascites that was present in the recipient twin had disappeared. By the 11th day after amniocentesis, the amniotic fluid in both sacs had normalized and remained as such until delivery at 35 5/7 weeks. The sudden resolution of the syndrome after one or more amniocenteses, and the failure to document an increase in urine production in the stuck twin in the few cases where it was measured, are not consistent with the explanation that the post procedural increase in the amniotic fluid around the stuck twin is always the direct result of improved fetal renal perfusion. We propose that, in cases where resolution of the syndrome occurred unexpectedly, the amnion separating the twins was inadvertently punctured at the time of amniocentesis. This resulted in movement of fluid along a hydrostatic pressure gradient from the sac with polyhydramnios into the stuck twin's sac. At first, it would seem implausible that the intervening membrane could be unintentionally perforated despite continuous ultrasonographic visualization of the needle within the sac with polyhydramnios. Figure 1, however, shows that the operator may be unaware that the intervening membrane is being traversed by the needle. Before one of the twins becomes stuck to one side of the uterus, the separating membrane floats freely between the two sacs. As the amniotic fluid increases in one sac, this membrane bulges toward the other sac and is ultimately pushed against the stuck twin and the uterine wall. Therefore, this membrane which is formed by the amnions of both twins does not only envelop the stuck twin, but also covers part of the internal uterine wall surface. Visualization of the intervening membrane along the inner uterine wall would not be possible given the limitations of our current ultrasound imaging equipment. This finding, however, was described by De Lia et al. during endoscopic laser treatment for twin-twin transfusion: "During the operation, it was apparent that the amniotic membrane septum between the twins had maintained its location over the vascular equator and was compressed onto the chorionic plate of the smaller twin by the expanding sac of the polyhydramniotic twin.". Since the membranes covering the uterine wall are not easily identified by ultrasound, one cannot be certain of the exact position of the separating membrane. When the needle happens to be placed over the uterine area covered by this membrane, as in position 1 in figure 1, both amniotic sacs will be punctured. If the needle placement happens to be in an area such as position 2 in figure 1, then only the amnion of the sac with polyhydramnios is punctured and no connection between the two sacs is created. We therefore hypothesized that the deliberate creation of a needle puncture in the intervening membrane, a procedure for which we coined the term septostomy, would allow equilibration of the fluid between the two sacs and improve survival of the twins without the need, in most cases, for decompression amniocentesis. The purpose of this report was to review the experience of five centers using this technique.

Figure 1	Figure 2