Percutaneous fetoscopic closure of large open spina bifida using a bilaminar skin substitute

OBJECTIVE We have described previously our percutaneous fetoscopic technique for the treatment of open spina bifida (OSB). However, approximately 20-30% of OSB defects are too large to allow primary skin closure. Here we describe a modification of our standard technique using a bilaminar skin substitute to allow closure of large spinal defects. The aim of this study was to report our clinical experience with the use of a bilaminar skin substitute and a percutaneous fetoscopic technique for the prenatal closure of large OSB defects. METHODS Surgery was performed between 24.0 and 28.9 gestational weeks with the woman under general anesthesia, using an entirely percutaneous fetoscopic approach with partial carbon dioxide insufflation of the uterine cavity, as described previously. If there was enough skin to be sutured in the midline, only a biocellulose patch was placed over the placode (single-patch group). In cases in which skin approximation was not possible, a bilaminar skin substitute (two layers: one silicone and one dermal matrix) was placed over the biocellulose patch and sutured to the skin edges (two-patch group). The surgical site was assessed at birth, and long-term follow-up was carried out. RESULTS Percutaneous fetoscopic OSB repair was attempted in 47 consecutive fetuses, but surgery could not be completed in two. Preterm prelabor rupture of membranes (PPROM) occurred in 36 of the 45 (80%) cases which formed the study group, and the mean gestational age at delivery was 32.8 ± 2.5 weeks. A bilaminar skin substitute was required in 13/45 (29%) cases; in the remaining 32 cases, direct skin-to-skin suture was feasible. There were 12 cases of myeloschisis, of which 10 were in the two-patch group. In all cases, the skin substitute was located at the surgical site at birth. In five of the 13 (38.5%) cases in the two-patch group, additional postnatal repair was needed. In the remaining cases, the silicone layer detached spontaneously from the dermal matrix (on average, 25 days after birth), and the lesion healed by secondary intention. The mean operating time was 193 (range, 83-450) min; it was significantly longer in cases requiring the bilaminar skin substitute (additional 42 min on average), although the two-patch group had similar PPROM rate and gestational age at delivery compared with the single-patch group. Complete reversal of hindbrain herniation occurred in 68% of the 28 single-patch cases and 33% of the 12 two-patch cases with this information available (P