Proceedings of the First Annual Meeting of the International Fetoscopic Myelomeningocele Repair Consortium

Abstract

Prenatal myelomeningocele (MMC) repair has significant advantages over postnatal repair, as was demonstrated by the Management of Myelomeningocele Study (MOMS) in 20111. This multicenter randomized controlled trial demonstrated that prenatal compared with postnatal repair significantly reduced the rate of ventriculoperitoneal shunt placement in the first 12 months of postnatal life and improved motor function, evidenced by a higher proportion of ambulatory infants at 30 months1. In MOMS, access to the fetal spinal defect was via a laparotomy with uterine exteriorization and hysterotomy, with consequent procedure-associated maternal and fetal risks, including increased risk of uterine rupture, need for Cesarean section in the index pregnancy and future pregnancies, and higher rate of preterm birth as well as perinatal mortality in future pregnancies1-3. Other complications more prevalent in the prenatal-repair group were: chorioamniotic membrane separation, spontaneous rupture of the membranes, oligohydramnios, placental abruption, pulmonary edema and need for maternal transfusion at delivery1, 2. Fetoscopic in-utero spina bifida repair was introduced with the objective of reducing the maternal and fetal/neonatal complications/risks of open hysterotomy repair, while at the same time preserving the neurologic benefits for the child4-14. Two different minimally invasive approaches have been developed and are currently being used in several centers worldwide: a totally percutaneous technique5-10 and an open fetoscopic technique on an exteriorized uterus11-14. After initial discussions about the different fetoscopic MMC repair techniques during the 26th Fetal Medicine Foundation World Congress in Slovenia in 2017, the idea of developing an international consortium for the study of fetoscopic MMC repair was originated by Kypros Nicolaides. He asked Michael Belfort, Denise Lapa and Elena Carreras to establish the Consortium and to facilitate a forum in which all groups performing fetoscopic MMC repair could work together transparently and collaboratively. The bylaws of the Consortium were established in December 2017 and a formal invitation to participate was sent out to all groups performing such surgery during the first half of 2018. The requirements for membership of the Consortium were then established by mutual agreement between the different centers that showed interest in being part of this initiative. These requirements include: first, submission of an approved protocol, providing a detailed description of the inclusion and exclusion criteria for surgery, the technique used and the plan for a minimum of 30 months of follow-up for patients who undergo fetoscopic MMC repair; second, submission of a copy of their Institutional Review Board or Ethics Committee approval letter; third, submission of an Investigational Device Exemption (IDE) approval letter from the Food and Drug Administration (FDA) for those centers located in the USA; fourth, provision of details regarding simulation training for their team, including, for example, the frequency of simulations, participants and systems; fifth, submission of a signed data sharing and confidentiality agreement; and sixth, performance of at least one case of fetoscopic MMC repair. The primary mission of the Consortium is to advance knowledge on fetoscopic MMC repair by fostering cooperative clinical research into this technique. To achieve this objective, the centers participating in the Consortium have formally agreed to provide accurate and complete data to a common registry. The data, including obstetric, perinatal, neurosurgical, neurodevelopmental and neurological outcomes, are modeled on the data structure of the MOMS trial1. These data will be reviewed and discussed by designated representatives of each participating center, with the stated intent of providing peer oversight, and the shared data will be presented transparently on an annual basis at a meeting held in conjunction with the World Congress in Fetal Medicine, under the auspices of The Fetal Medicine Foundation. Participants were asked to present a summary of their surgical results and related studies. For centers with greater experience and those in which the surgical technique had undergone different iterations, representatives were asked to present their center data excluding the ‘learning curve’. The main points from the presentations are given in Appendices S1 and S2 and are summarized below. The results from centers that had, at the time of the meeting, performed five or more procedures with delivery are summarized in Tables 1-3. Centers that were not yet performing fetoscopic MMC repair surgery but were interested in developing this approach at their institutions (listed in Appendix S3) were also invited to this meeting and participated in the discussions. Dural +/– artificial skin Dural +/– artificial skin Total 80 Learning curve 20 Results of last 60† 39/58 (67) 3/8 (38)¶ 39/58 (67) 3/8 (38)¶ 53/58 (91) 2/8 (25)¶ 11/58 (19) 0/8 (0)¶ Total 45 Learning curve 33 Results of last 10‡ 2017– 2018 Total 38 Learning curve 29 Results of last 9 Total 14 Learning curve 0 Results of last 11§ Total 8 Learning curve 0 Results of all 8 Total 5 Learning curve 0 Results of all 5 26/29 (90) (≥ 12 months) 13/37 (35)** 4/14 (29)¶ 10/14 (71)¶ 1/8 (13); 1/3 (33) 2/10 (20); 2/6 (33) 0/8 (0); 0/1 (0) 3/4 (75); 1/1 (100) This procedure, developed by Dr Denise Lapa7, essentially involves the following steps: first, general anesthesia is given; second, ultrasound-guided amnioinfusion of 500 mL warmed saline/ringer is performed; third, three 11-Fr and one 5-mm balloon-tipped laparoscopic trocars are inserted percutaneously; fourth, almost all amniotic fluid is removed, and heated and humidified carbon dioxide (CO2) is insufflated into the uterus; fifth, an endoscope, graspers, scissors and needle driver are passed through the ports; sixth, the placode is released and the skin edges are undermined to allow their approximation in the midline; seventh, a dural patch is placed over the defect and then a bilateral aponeurosis flap is sutured in the midline; eighth, the skin is closed. In cases in which skin approximation is not possible, a bilaminar skin substitute is placed over the dural patch and sutured to the skin edges. Fetal surgery by a totally percutaneous fetoscopic technique was carried out on 80 cases, but data from only the last 60 consecutive cases were presented since the first 20 cases were considered to be part of the learning curve and had been published previously7. The inclusion criteria were neural tube defect located at any level, provided the parents understood the severity of the defect, gestational age at surgery between 24.0 and 28.9 weeks, presence of hindbrain herniation, no other major abnormalities and normal karyotype. Excluded were patients with placenta previa, alloimmunization, multiple gestation, positive serology for HIV or hepatitis B or C, and maternal conditions increasing the risk for surgery or anesthesia (such as uncontrolled diabetes or hypertension). Insufflated CO2 was heated and humidified in only the last eight survivors. The median (range) gestational age at the time of surgery was 26.6 (24.0–28.9) weeks; there were 58 live births, one fetal death and one pregnancy termination. The median gestational age at delivery was 32.5 (26.9–40.7) weeks, and the rate of preterm prelabor rupture of the membranes (PPROM) was 67% (39/58). Of the 58 survivors, five (9%) had cerebrospinal fluid (CSF) leakage and six (10%) required revision of the repair at birth. At the time of reporting, 30 children were ≥ 12 months of age, of which 14 (47%) required a ventriculoperitoneal shunt or third ventriculostomy. Follow-up at ≥ 30 months of age found that 59% (10/17) of infants were ambulatory. Bladder function was assessed in 37 of 58 survivors at a median age of 22 (range, 1–60) months and in 24 (65%) there were no signs of neurogenic bladder; none of these children was taking medication for bladder dysfunction or infection, and none required intermittent catheterization. It was concluded that large MMC defects may be treated successfully in utero using a totally percutaneous fetoscopic technique, but that PPROM and preterm birth remain a challenge; the incidence of these complications is likely to be reduced by the introduction of heating and humidification of the insufflated CO2; among the last eight cases analyzed (excluding one termination), the median gestational age at birth increased to 34.5 weeks and the rate of PPROM decreased to 38% (3/8). Fetal surgery was reported for four cases; the first was performed by Dr Lapa with the assistance of the team of Dr Gielchinsky, and the second by Dr Gielchinsky with his team, under the supervision of Dr Lapa and her team. The subsequent two cases were performed independently by Dr Gielchinsky's team. The inclusion and exclusion criteria and technique were essentially the same as those of Dr Lapa, described above. After removal of amniotic fluid, heated CO2 was insufflated into the uterus without the use of a humidifier. The median (range) gestational age at the time of surgery was 25.7 (24–28) weeks and at delivery was 32.3 (30–33) weeks. All cases had PPROM, at a median (range) gestational age of 30 (28–32) weeks. All babies survived and there were no cases with CSF leakage or requiring revision of the repair at birth. Two (50%) babies underwent ventriculoperitoneal shunt placement within the first 12 months. Fetal surgery was reported for two cases; the first was performed by Dr Lapa with her team assisting, and the second was performed by Dr Yamamoto's team under the guidance of the teaching team. The inclusion and exclusion criteria and technique were essentially the same as those of Dr Lapa, described above. After removal of amniotic fluid, CO2 was insufflated into the uterus without heating or humidification. Surgery was performed at 27 weeks' gestation, PPROM occurred at 29 weeks and delivery occurred at 33 and 36 weeks, respectively. Neither case had CSF leakage at birth and neither required postnatal repair. Postnatal ventriculoperitoneal shunt placement was carried out in one of the two cases. Fetal surgery was reported for one case; this was performed by Dr Simpson's team under the guidance of Dr Lapa. The technique was essentially the same as that of Dr Lapa, described above. Surgery was performed at 27.3 weeks, PPROM occurred at 28.5 weeks and the baby was delivered at 29.2 weeks. There was no CSF leakage at birth and the baby did not require postnatal repair. A temporary ventricular subgaleal shunt was placed in the neonatal period and was removed subsequently. This procedure essentially involves the following steps: first, general anesthesia is given, laparotomy is performed, the uterus is exteriorized and the fetus is gently manipulated into position using ultrasound guidance; second, either two or three 10–12-Fr cannulas are introduced between the uterine wall and amniotic cavity (in some centers, plication sutures are placed to fix the membranes to the uterine wall prior to insertion of the cannulas); third, amniotic fluid is withdrawn and CO2 is insufflated into the uterus (in some centers, the CO2 is heated and humidified); fourth, an endoscope, graspers, scissors and needle driver are passed through the ports; fifth, the placode is dissected; sixth, in centers using single-layer closure, the skin is sutured over the open defect, in centers using double-layer closure, a dural patch is placed over the defect before suturing the skin, and in centers using triple-layer closure, a dural patch is placed over the defect and then durafascial or myofascial flaps are cut parallel to the spinal defect and sutured in the midline before suturing the skin. If needed, relaxing skin incisions (15–20 mm lateral to the defect) are created to allow midline apposition of the skin edges. Fetal surgery was reported for 45 cases, but the focus was on the last 12 patients, in whom a two-port technique with triple-layer closure and placement of a dural patch was used. The inclusion and exclusion criteria were the same as those in the MOMS trial1. The membranes were plicated to the uterine wall before port insertion. Insufflated CO2 was heated and humidified. In the 12 cases, the median (range) gestational age at surgery was 25.0 (24.0–26.0) weeks. cases had been delivered by the time of this and their median (range) gestational age at delivery was weeks. The PPROM rate was All babies were case had repair or leakage of CSF at birth. This to be a major of the triple-layer closure, in the of single-layer the CSF rate was of the children was ≥ 12 months of age at the time of (10%) case required a ventriculoperitoneal shunt during the first follow-up was presented for the of none of the patients had 30 months of age for of 14 cases at four were to independently and with a neurodevelopmental patients underwent and with at months and at with the proportion of abnormalities in different from to in the and to in the were also presented which showed first, the use of for and of following fetoscopic repair is and second, fetal and postnatal fetoscopic MMC repair is not significantly different from that MMC repair in the same Fetal surgery was reported for cases. The inclusion and exclusion criteria were the same as those in the MOMS trial1. A technique with double-layer closure was used in the last two patients, a dural patch was The membranes were plicated to the uterine wall before port insertion. In the last two cases the insufflated CO2 was heated and humidified. The median (range) gestational age at surgery was weeks and at delivery was weeks. There were eight survivors and one fetal The PPROM rate was 38% (3/8). case had repair or leakage of CSF at birth. three of the eight children were ≥ 12 months at the time of and one of required a ventriculoperitoneal Fetoscopic surgery was reported for 14 cases, three of whom were at the time of The inclusion and exclusion criteria were the same as those in the MOMS trial1. A or technique with or triple-layer closure with placement of a or dural patch over the placode and skin closure, with or without a skin over the was used. were not plicated before port insertion but the port were with sutures Insufflated CO2 was The median (range) gestational age at surgery was weeks and at delivery was weeks. There were live and one fetal death with a The PPROM rate was case had repair or leakage of CSF at birth. six of the children were ≥ 12 months at the time of and 2/6 required a ventriculoperitoneal shunt at 9 months and the other at months of Fetal surgery was reported for eight cases. The inclusion and exclusion criteria were the same as those in the MOMS but with an inclusion of of A two-port technique with single-layer closure and no patch was used. The membranes were plicated to the uterine wall before port insertion. In the last three cases, the insufflated CO2 was heated and humidified. The median (range) gestational age at surgery was weeks and at delivery was weeks. The PPROM rate was and cases were treated prior to the use of humidified CO2 case had repair or leakage of CSF at birth. one of the eight children was ≥ 12 months at the time of and that had not required a ventriculoperitoneal Dr also reported in a study of a of four of fetal at the and of surgery, there were no that to of CO2 did not fetal Fetal surgery was reported for five cases. The inclusion and exclusion criteria were the same as those in the MOMS but with of the gestational age to weeks. A technique with single-layer closure and no patch was used. The membranes were plicated to the uterine wall before port insertion. The insufflated CO2 was not heated or humidified. There were four survivors and one fetal The median (range) gestational age at surgery was weeks and at delivery was weeks. The PPROM rate was In two there was CSF leakage at birth and one revision of the repair. In three of the four a ventriculoperitoneal shunt was during the first postnatal Dr their results were those in the MOMS the technique may be postnatal repair in their He some for including heating and humidification of CO2 for and use of a dural patch to postnatal Fetal surgery was reported for two cases. The inclusion and exclusion criteria were the same as those in the MOMS trial1. A two-port technique with triple-layer closure was to that described by Dr The membranes were plicated to the uterine wall before port insertion. and humidified CO2 was used. Surgery was performed at 24 weeks' in one case and weeks in the and PPROM by delivery occurred at and 33 weeks, respectively. At in one case there was leakage of CSF requiring repair and in the second there was no The first baby required ventriculoperitoneal and the second had a temporary subgaleal The of the first meeting of the Fetoscopic MMC Consortium participated in a a of and stated for the to These are discussed below. All participants agreed that the of the Consortium under the auspices of The Fetal Medicine Foundation is and all to their data to the in a and The bylaws of the Consortium were discussed and agreed All groups agreed that fetoscopic MMC repair surgery some of the significant maternal related to the These or of the hysterotomy uterine rupture and need for multiple of the groups had of the port access in patients who had a Cesarean which with the rate of hysterotomy reported in the MOMS trial1. centers performing fetoscopic repair the of which is a of the fetoscopic approach is the for a delivery without in a neonatal This has and benefits for the The neurological on to those from the MOMS trial1. The experience reported by Dr Lapa that the incidence of neurogenic bladder may be that reported by the MOMS to some could be to the patch used in the totally percutaneous technique. In of follow-up at months may the such as the need for surgery to the is only 30 is likely that will be reduced the the from the this is by data, different and and the use of different and of Two surgical techniques to access to the fetus for fetoscopic repair of MMC were by the a percutaneous approach and an there are of each technique that have been by different including in the of use of surgical and there two on the use of an a percutaneous It was agreed that there is an need for all to the same neurological and in a The groups that of cases are needed, with and of outcomes, before any be over the The of each of the two techniques stated data on all be compared be to their approach to is in an to be This be one of the of the of the technique of fetoscopic repair, there is a rate of membrane and There are a of for of the membranes at the time of port of the membranes by the of CO2 on the membranes, to the membranes during surgery, of the uterus by CO2 of the uterus and fetus during surgery, of the amniotic fluid by and by of the uterine to the or of these is is to humidification and heating of the CO2 to the of of the membranes, as well as plication of the membranes to the uterine wall prior to port insertion. There was general of the need for all to the use of heated and humidified There was general by the participants that the of CO2 use is for in to CO2 between and were discussed and the different and was a the experience from over cases that fetal CO2 is not with any perinatal The rate of fetal was with that reported in the MOMS trial1. on the experience of the participants with more experience and to be to that in the MOMS trial1. data will be to more patients in of and These results are of the to this The need for a approach to achieve a closure was also and data are and is The use of was with of regarding the use of relaxing incisions as the to achieve a closure in a or large MMC The of relaxing incisions that an skin closure over the spinal defect was to an open that by under a reduced the risk for neonatal and provided of the The of were of the large defects that from relaxing incisions and stated that neurological with a bilaminar patch are and neonatal data will this but there is an need for the of and from open at and the of the defect over time well as any such as and of this on the of used in the There was general agreement that myofascial flaps be but there was no agreement on the need to the placode with a patch prior to the The team from presented data significant in following their of a triple-layer closure technique that the use of myofascial flaps sutured over a patch by skin The use of sutures sutures for skin closure was also with no The for of a fetoscopic MMC repair has a common of the technique and first, a team, of and and on the need for such a second, for the third, to an center to one or more cases; and fourth, in are to their first centers will for one or more of the team approach are to be present to and their The need for more a surgical team was It was agreed that any this surgery have the neonatal and as well as a spina bifida and and other to the and of these Participants agreed that the is a and to in to this surgery in The of training of the team, using different simulation is for the of their It is for this that of the use of a simulation in each of the participating centers is one of the requirements for in the participants on the of of all in to and performance of this This was agreed as an to at the same The procedure only be performed under the of an Ethics Committee or Institutional Review but participants were to of oversight, such as a Review Board Fetal All be on and all participating in the have an is that will all the Consortium and their data and in a and accurate All currently participating have to peer of their data by a group of representatives from each In this an be The main of this meeting the following general points of agreement by first, regarding the requirements to part of this Consortium and the and that such membership second, to their data to the common and to and peer of the a for the of a fetoscopic MMC repair technique be at this all in the group that is one of the main of this data from the different centers performing the different techniques any of each is in to this The with the experience are to their knowledge and agreed to and sharing may be through simulation clinical and research as well as through at the Consortium Denise Medicine, Surgery and Michael and and and and of Medicine, Elena and and and and Medicine and and and Fetal and and Fetal Fetal and and Fetal Medicine and for Fetal and Fetal and and Medicine Gielchinsky's of and of and Fetal and Yamamoto's and Simpson's Fetal Appendix S1 of percutaneous fetoscopic approach Appendix S2 of fetoscopic approach Appendix of centers interested in fetoscopic myelomeningocele repair The is not for the or of any by the be to the for the