Twin reversed arterial perfusion ( TRAP ) sequence – does monoamniocity preclude early intervention?

A 27-year-old primigravida was referred at 13 weeks' gestation with twin reversed arterial perfusion (TRAP) sequence. The initial scan showed a monoamniotic twin pregnancy including an anencephalic twin without cardiac activity, with retrograde perfusion by an apparently normal pump twin without signs of cardiac compromise. Abdominal circumference and femur length were the same in each twin. The three-vessel cord of the pump twin and the two-vessel cord of the TRAP twin inserted into the placenta with a distance of 15 mm and were twisted together without a knot (Figure 1). The parents were counseled on the different options (expectant management with late intervention in case of cardiac compromise of the pump twin or early intervention by intrafetal laser occlusion), and opted for early intervention. On the next day intrafetal laser occlusion was performed as previously described1. Subsequent ultrasound examinations at 2-week intervals showed the expected decrease in size of the unperfused TRAP twin with undisturbed umbilical flow in the pump twin on color-flow and pulsed-Doppler ultrasound. The pregnancy continued uneventfully until 22 weeks' gestation, when intrauterine demise of the pump twin was diagnosed. Labor was induced and the specimen showed ligation of the pump twin's cord by one loop of the string-like umbilical cord of the TRAP twin (Figure 2). Ultrasound-guided intrafetal laser is a simple and widely available technique that has been successfully applied in the treatment of TRAP sequence in early pregnancy, when fetoscopy is not applicable. In a recent meta-analysis of 18 interventions performed before 16 weeks' gestation an overall neonatal survival of 88.9% was reported1. In our own recent study the loss rate following intrafetal laser therapy was two of eight cases (25.0%) performed before 16 weeks. All cases without intrauterine death continued uneventfully until term2. However, none of these interventions was performed in monoamniotic TRAP sequence. There is an ongoing discussion as to whether cord entanglement itself complicates monoamniotic pregnancies. In a prospective observational study Dias et al.3 investigated the outcome of 32 monoamniotic twin pregnancies: they found entanglement in all cases and concluded that perinatal mortality is mainly a consequence of conjoined twins, TRAP, discordant anomaly and spontaneous miscarriage before 20 weeks' gestation. The seven TRAP pregnancies in this series ended in three miscarriages and four live births subsequent to intrafetal laser therapy. This suggests that once the circulation of the TRAP mass has been disrupted, the prognosis is favorable, despite monoamniocity and cord entanglement. In contrast, Hecher et al.4 reported two cases of late intrauterine death with apparent cord entanglement after successful cord occlusion in monoamniotic TRAP sequence. The true cause of death of the pump twin in untreated TRAP sequence with cord entanglement may be a consequence of hemodynamic disturbances rather than from cord-related complications. However, what mechanism other than strangulation on the basis of entanglement could be the cause of intrauterine death weeks after successful occlusion of the TRAP circulation? Furthermore, it seems reasonable that a string-like umbilical cord of a dead twin is more likely to cause complications. These considerations have resulted in the recommendation of performing cord occlusion and cord transection in selective reduction or TRAP sequence in monoamniotic twins5, 6. Clearly, with the instruments available today, these measures cannot be performed in the first trimester. Our case demonstrates that, even weeks after successful intrafetal laser coagulation of a monoamniotic TRAP twin in the first trimester, strangulation of the umbilical cord of the pump twin might occur. Although the incidence of this type of complication is unknown, it seems reasonable to avoid early intervention in monoamniotic TRAP sequence in the absence of hemodynamic compromise of the pump twin, and to postpone treatment until cord transection is feasible. Consequently, cord transection has to be considered even after spontaneous cessation of perfusion to a monoamniotic TRAP twin. We thank Lisbeth Lewi for her most valuable comments on this case. C. Berg*†‡, A. Köninger§, U. Gembruch‡ and A. Geipel‡ †Division of Prenatal Medicine and Gynecologic Sonography, Department of Obstetrics and Gynecology, University of Cologne, Cologne, Germany; ‡Division of Fetal Surgery, Department of Obstetrics and Prenatal Medicine, University of Bonn, Bonn, Germany; §Division of Obstetrics and Perinatal Medicine, Department of Obstetrics and Gynecology, University of Essen, Essen, Germany *Correspondence. (e-mail: [email protected])