Percutaneous minimally-invasive fetoscopic surgery for spina bifida aperta – Part I Surgical technique and perioperative outcome Thomas Kohl MD German Center for Fetal Surgery & Minimally Invasive Therapy (DZFT), University of GiessenMarburg, Germany
Corresponding author: Thomas Kohl MD German Center for Fetal Surgery & Minimally Invasive Therapy (DZFT) University Hospital Giessen-Marburg Klinikstr. 33 35592 Giessen – Germany
phone: +49-175-597-1213 fax: +49-641-985-45279 [email protected] www.dzft.de
Keywords fetoscopy, fetal surgery, spina bifida, Chiari-II-malformation, hydrocephalus, fetus
This article has been accepted for publication and undergone full peer review but has not been through the copyediting, typesetting, pagination and proofreading process, which may lead to differences between this version and the Version of Record. Please cite this article as doi: 10.1002/uog.13430
This article is protected by copyright. All rights reserved
Abstract Background Percutaneous minimally-invasive fetoscopic closure of spina bifida aperta (SBA) was developed by the author in order to obviate the need for the far more invasive open fetal surgical approach. The manuscript offers an analysis of the current technical approach accompanied by an orienting overview on its development in ovine and human fetuses. Flanked by supporting manuscripts about management and outcome, the author hopes to raise the interest in the further dissemination of minimally-invasive surgery for this condition. Patients & Methods Minimally-invasive percutaneous fetoscopic closure of SBA was performed at the German Center for Fetal Surgery & Minimally-Invasive Therapy (DZFT) in 51 human fetuses between 21.0 and 29.1 weeks of gestation (mean age 23.8 weeks). Various parameters of surgical relevance for the success and safety of the procedure and the early perioperative outcome were retrospectively analyzed. In addition, historical information from the early clinical cases and how it shaped the mature approach is is being discussed. Results Percutaneous minimally-invasive fetoscopic closure of SBA was performed with a high rate of technical success, regardless of placental or fetal position. Fetal demise one week after the procedure occurred in one of the 51 cases (2%) from early postoperative chorioamnionitis. All other fetuses survived gestation. Of the surviving fetuses 45 (90%) were born beyond 30 weeks and 25 (49%) beyond 34 weeks of gestation. Conclusions Following an adequate learning curve, minimally-invasive fetoscopic surgery for fetal spina bifida be performed with a high rate of technical success regardless of placental position.
This article is protected by copyright. All rights reserved
Introduction Open fetal surgery for spina bifida aperta (SBA) has been performed in hundreds of human fetuses. The Management-Of-Myelomeningocele-Study (MOMS) provided level I - evidence that, compared to prenatally unoperated fetuses, prenatal SBA closure may preserve leg function and reduce the severity of hindbrain herniation and hydrocephalus in affected fetuses (1). Yet, the open approach is associated with significant maternal morbidity as it requires maternal laparotomy and hysterotomy for fetal access. In a significant proportion of cases, the hysterotomy scar left after open fetal surgery becomes a weak spot, prone to uterine wall dehissence or even rupture after fetal surgery or in future pregnancies.
As the clinical consequences of SBA provide multiple reasons for life-long disabilities, sorrows and requirements for therapy, the maternal trauma from the open approach has been accepted by many expectant mothers and prenatal specialists as the prize to be paid for the clinical benefits. Yet, even the earliest prenatal procedures in the mid-1990s already aimed at reducing maternal injury: In order to avoid maternal hysterotomy, Drs. Bruner and Tulipan performed their pioneering surgeries by fetoscopy, securing skin grafts to the malformation (2,3). Their technique, however, still required maternal laparotomy followed by transuterine trocar placement. It was quickly abandoned in favor of the open approach because of technical difficulties and unfavorable outcomes. A few other attempts at fetoscopy were performed at UCSF but subsequently also abandoned for similar reasons (4).
During studies in inanimate models, sheep and postmortem human fetuses, the author developed an even less invasive because fully percutaneous minimally-invasive fetoscopic approach for prenatal closure of SBA (5-10). This manuscript offers an orienting overview of its development and presents technical data as well as perioperative outcomes from 51 cases that were performed at the German Center for Fetal Surgery & Minimally-Invasive Therapy (DZFT) over the past three years. Flanked by accompanying manuscripts about selection, perioperative management and the mostly promising neonatal and 30-month neurological outcomes, the author wishes to promote the dissemination and further development of minimally-invasive fetoscopic surgery for SBA.
This article is protected by copyright. All rights reserved
Patients & Methods Various parameters of surgical relevance for the success of percutaneous minimally-invasive fetoscopic spina bifida closure were retrospectively analyzed in 51 cases that were performed at our center between July 2010 to June 2013 (Table). Maternal age at surgery ranged between 19 and 41 years with a mean age of 31.2 years. Gestational age at surgery ranged between 21.0 and 29.1 weeks of gestation with a mean age of 23.8 weeks.
Analysis of this data has been approved by the local committee on human research (#158/13). The procedure has been offered as standard of care at the German Center for Fetal Surgery & MinimallyInvasive Therapy (DZFT) since July 2010 following 1. completion of a CHR-approved pilot study in 30 patients at the University of Bonn, Germany, 2. when an independent match-controlled study in 13 of these pilot study patients showed that postnatal leg function was significantly better and the rate of shunt-dependant hydrocephalus was significantly lower than in prenatally unoperated controls (11), and 3. when the technical success rate, as well as maternal and fetal safety, and gestational age at delivery had been greatly improved upon compared to the early pilot phase experience.
Fetal selection The gestational age of the fetuses undergoing percutaneous fetoscopic patch coverage of SBA preferably ranges between 20+0 and 25+6 days. Only a few procedures are carried out later when the degree of hydrocephalus is still mild and stool soiling of the lesion can be anticipated from characterization of ist surgical anatomy by ultrasound (12,13). The defect should be located between Th1 to S1. Fetuses with a gibbus are excluded. Further requirements are hindbrain herniation and a diameter of the lateral brain ventricles of less than 16 mm as assessed by ultrasound and MRI. The fetuses should be free from other major organ abnormalities on ultrasound and present a normal karyotype. Yet some expectant women, following counseling, elect to forgo preoperative amniocentesis. In these cases, fetal surgery is offered when ultrasound- and MRI-studies do not raise
This article is protected by copyright. All rights reserved
suspicions of any other major fetal malformations or chromosomal anomalies. Following this selection policy, one fetus, whose parents had opted against preoperative karyotyping, that underwent technically successful surgery and was born at 36 weeks of gestation, died from trisomy 13 after delivery. This case was excluded from this analysis.
Intraoperative ultrasound-monitoring Apart from providing the fetal patient in the first place, multimodal ultrasound provides information critical for the safety and success of minimally-invasive fetoscopic closure of SBA: Surgically relevant information about the lesion that helps in planning the procedure, selecting safe and suitable trocar sites during intraamniotic access, monitoring of uterine closure and last but not least hemodynamic information is collected. Doppler ultrasound interrogation of the fetoplacental circulation as well as determination of fetal heart rate and contractility are regularly performed at defined times: before and after induction of anesthesia, after providing the mother with central venous and arterial lines, before intraamniotic access, after amnioinfusion, before insufflation and at the end of the procedure. This hemodynamic information helps the anesthesiologist to adjust the depth of anesthesia and to optimize maternal and fetal hemodynamics by maternal administration of volume, catecholamines and other drugs to the requirements of the procedure in dosages that warrant safety of both. Only when maternal and fetal hemodynamics remain stable after induction of anesthesia and, if required for achieving sufficient intraamniotic work space, amnioinfusion, the first trocar will be placed. If maternal or hemodynamics or fetoplacental blood flow deteriorate after induction of anesthesia or amnioninfusion, and attempts at their normalization are in vain, a procedure will be abandoned. Fortunately, this situation is rare and has been encountered only once over the past three years.
Surgical approach Percutaneous minimally-invasive fetoscopic surgery for SBA is performed via three (to four) trocars with an external diameter of 5 mm that are placed into the amniotic cavity under continuous ultrasound monitoring by a Seldinger approach (Fig 1). Following partial removal of the amniotic fluid, partial amniotic carbon dioxide insufflation (PACI) is performed as previously described in order to
This article is protected by copyright. All rights reserved
maintain visibility of intraamniotic contents (4,6). Fetal posturing is then carried out using fetoscopic instruments (Karl Storz, Tuttlingen, Germany). Then, the lesion is being circumcised with a needle electrode and the neural tissue is carefully freed from surrounding tissues (Fig 2 & 3). Depending on the anatomy of the lesion, the neural cord is covered with one or more patches. Regardless of the closure technique, the goal of watertight coverage is demonstrated at the end of the procedure by observing bulging of the patch as well as lack of cerebrospinal fluid leakage when its surface is being compressed with an instrument. Once the lesion has been closed, the insufflated carbon dioxide is being removed during simultaneous refilling of the amniotic cavity with a warm sterile crystaline solution. Following percutaneous coverage of the membranous defects at the trocar insertion sites, the abdominal trocar insertion sites are closed with non-absorbable 5-0 surgical sutures.
Study variables The following parameters were analyzed: Gestational age at delivery, acute mortality of surgery, placental position, success of percutaneous fetal access, number of trocars, incidence and consequences of trocar dislodgment, success of fetal posturing, pressures and duration of partial amniotic carbondioxide insufflation (PACI), need for additional venting of insufflation gas from the maternal abdomen, type of lesion (flat, cystic), lesion size, fetoscopic handling of the lesion (patch closure, direct closure, combination of both), overall technical success of the procedure defined as patch coverage of the neural tissue, success of uterine closure, and skin-to-skin-time of fetoscopic procedure.
Results Gestational age at delivery in survivors ranged between 27.9 and 38.1 weeks of gestation with a mean age of 33.0 weeks. All fetuses survived surgery and all but one gestation. The fetus that died, died from preterm delivery due to chorioamnionitis at 24.6 weeks about one week after the procedure (Case 5; Table). During the first months of life, two more infants died from severe brain-stem dysfunction attributed to the Chiari II – malformation (Cases 3 & 7; Table).
This article is protected by copyright. All rights reserved
20 procedures were performed in the presence of an anterior and 31 procedures in the presence of a posterior placenta. Skin-to-skin time ranged between 45 and 315 minutes. The 45-minute-case was abandoned because maternal obesity in combination with an anterior placenta made completion of the procedure technically impossible. The shortest successful case took 140 minutes in a fetus with a flat lesion that could be easily dissected and covered by a single-patch. Thirteen procedures were performed between 140 to 200 minutes, 25 procedures took between 201 – 250 minutes. Eleven procedures were performed between 251 and 301 minutes. The longest case took 315 minutes. In this case, the surgical anatomy was extremely complex and triple patch coverage had been performed. Ultrasound-guided percutaneous intraamniotic access with ports with an external diameter of 5 mm was established in all cases. During 49 procedures three and during two procedures four trocars (Cases 5 & 7; Table) were inserted into the amniotic cavity. In cases 5 and 7, the additional trocar was required for maintaining a stable fetal position throughout the procedure. In one case, dislodgment of one trocar occured at the end of one procedure such that closure of the trocar insertion site could not be performed (Case 24; Table). In another case that had to be abandoned, dislodgment of two trocars occurred right after induction of insufflation as a consequence of maternal obesity (Case 32; Table). Favorable percutaneous fetal posturing was achieved in all but the one case where early dislodgment of trocars occurred. Opening pressures for PACI ranged between 9 to 30 mm Hg (mean 15.6 mm Hg) and duration ranged between 10 – 275 minutes (mean 183 min; Table). PACI was well tolerated by both mother and fetus and permitted clear visualization throughout the entire fetoscopic procedure. Additional venting of insufflation gas from the maternal peritoneal cavity was required in 17 cases. There were 35 flat and 15 cystic lesions. Measured at skin level, their areas ranged between 3 to 12 cm2. Overall technical success of the procedure defined as complete patch coverage of the exposed neural tissue at the end of fetal surgery was achieved in all but the one case where early dislodgment of two trocars occurred in an obese patient (Case 32; Table). Single-patch covererage was performed in 30 fetuses, double-patch coverage in 14 (Fig. 2), triple-patch coverage in two and direct closure of the lesion in four fetuses. In two fetuses of the latter group, the skin was closed immediately above the neural cord without interposition of patch material (Cases 35 & 44; Table). In the two other ones,
This article is protected by copyright. All rights reserved
and recommended by our local neurosurgeon, direct skin closure was performed after covering the neural cord with a collagen patch (Cases 48 & 51; Table; Fig. 3). Coverage of all membranous defects at trocar removal was achieved in 47 cases, coverage of only two of three access sites was achieved in the remaining 4 cases. The reasons for not achieving or forgoing coverage were trocar dislodgment in two cases, avoiding the risk of umbilical cord compromise in one case and two too closely situated trocar insertion sites in another one.
This article is protected by copyright. All rights reserved
Discussion This retrospective analysis shows that percutaneous minimally-invasive fetoscopic surgery for SBA in human fetuses can be performed with a high rate of technical success regardless of placental position, and with a low rate of perioperative mortality. Like the open fetal surgical approach, percutaneous fetoscopic patch coverage of the malformation aims at protecting the spinal cord tissue, reversing hindbrain herniation and reducing the need for postnatal ventriculo-peritoneal shunt insertion. Compared to open fetal surgery for SBA, however, the percutaneous minimally-invasive fetoscopic approach offers a major reduction of maternal injury by avoiding laparotomy and hysterotomy. Hence, maternal pain and discomfort are usually minimal beyond the 2nd day after surgery. In addition, most expectant women can be discharged home from hospital within a week. Adhering to the fundamental ethical principle of primum non nocere, further exploration and dissemination of the minimally-invasive fetoscopic techniques seem worthy goals and may over time allow replacing the open operative approach. Employing ultrasound-guidance, percutaneous fetal access with three trocars can be achieved in essentially all cases. Rarely, a fourth trocar is required for securing a favorable fetal lie throughout the procedure. As each trocar is inserted over a wire guide, the intial puncture hole for maternal percutaneous-transabdominal-transuterine-intraamniotic trocar insertion has a diameter of only 1.2 mm. This gentle method of intraamniotic access is the hallmark of percutaneous minimally-invasive fetal surgery and is barely comparable to the much larger incisions in the maternal abdomen and uterus required for open fetal surgery (Fig. 4). Whereas I assumed in the beginning that percutaneous minimally-invasive fetoscopic spina bifida closure might only be feasible in women with posterior, lateral or fundal placentae, growing clinical experience showed that the operation can be performed in most cases regardless of placental position. During ultrasound-guided trocar insertion in cases with an anterior placenta, a safety margin of about 3 cm away from the placental edge suffices in preventing placental injury. Trocar dislodgement resulting in loss of surgical setup has been fortunately rare in this clinical series. It usually suffices to advance each trocar for about 3-5 cm into the amniotic cavity. Yet, as seen in two of our cases, trocar dislodgment may occur during instrument manipulations or in adipose patients in whom the trocar tip can sometimes be advanced for only 2 cm or less into the amniotic cavity. Trocar dislodgment can also be the consequence of too much traction between the abdominal
This article is protected by copyright. All rights reserved
and uterine walls or of sudden uterine volume changes from intraoperative uterine contractions or when the amniotic cavity gradually collapses from unattended gas loss along the trocars into the maternal abdomen (5). In cases with an anterior placenta, trocar dislodgment requires to abandon the procedure since amniotic gas insufflation precludes ultrasound-guided reinsertion of another trocar. Unmonitored or in the presence of impaired ultrasound imaging quality, trocar insertion in cases with an anterior placenta carries a high risk for placental or maternal bowel injury. In contrast, in cases with a posterior placenta, reinsertion can easily be achieved through the original opening guided by fetoscopy. Successful fetal posturing with endoscopic instruments can usually be achieved in most cases. Yet, the more obese a patient, the more difficult posturing maneuvers can become. In this situation, the fragile instruments and endoscopes oftentimes bend to critical degrees. In obese patients with a posterior placenta postponing the procedure until the fetus lies in a more favorable position has been a helpful strategy. Alternatively, when more severe obesity comes with an anterior placenta, maternal laparotomy and transuterine trocar placement or even open fetal surgery may be considered. Due to the small trocar size, continuous pressure- and volume-controlled exchange of large volumes of crystalline solutions in order to manage bleeding complications or provide a clear field for dissecting and closing the malformation throughout the entire procedure may be difficult or impossible to achieve. In order to overcome the disadvantages of operating in fluid, management principles and safety measures for percutaneous gas insufflation of the amniotic cavity were defined from animal studies in sheep and early clinical experience (5,6,7,14,15). PACI over a mean period of three hours permitted clear visualization throughout all 51 fetoscopic procedures and did not result in any critical or unmanageable maternal cardiovascular effects as assessed by arterial blood gas analyses, continuous maternal ECG, blood pressure, oxygen saturation and end-expiratory carbon dioxide concentration monitoring. Encouragingly, postnatal clinical examinations as well as ultrasound- and MRI-scanning in the infants did not reveal any evidence of chronic brain injury that could be attributed to the fetoscopic setup. These clinical findings were supported by previous insufflation studies of our group in sheep (15,16). During the earliest clinical cases, it was entirely unclear how well and for how long mother and fetus would tolerate anesthesia and in particular PACI. At this time, my attempts at securing funding for amniotic insufflation studies in primates with the goal of studying various issues relevant to maternal
This article is protected by copyright. All rights reserved
safety of the novel technique were in vain. Years earlier, Bruner et al. were the first who insufflated an exteriorized uterus in humans during four fetoscopic procedures and encountered placental abruption in one of them (2). In contrast to their approach and making matters more complicated, PACI had never been used before in a percutaneous fetoscopic setup in humans. Further discouraging, previous studies of other investigators on exteriorized sheep suggested that carbon dioxide and/or insufflation pressures higher than 15 mm Hg would lead to dangerous degrees of fetal acidosis (1719). During own studies in sheep (a species that has a syndesmochorial placenta), I observed that insufflation of the amniotic cavity was tolerated at much higher insufflation pressures than previously thought without interfering with fetoplacental blood flow when the maternal abdomen was closed. Nevertheless, the first human procedures in this uncharted territory were kept as short and as simple as possible. Whenever a minor bleeding complication or some membrane detachement was detected, the procedure was abandoned for maternal safety considerations. As a result of that caution, three of the first 19 procedures went uncompleted (11).
In order to keep the procedure short in the initial cases, I simply covered the lesion using inert patch material and left the malformation itself untouched (9). Due to the strong traction forces on the patch rim from fetal growth, the patches often partially dehissed and coverage of the entire lesion until delivery was seldom achieved. As a consequence, during this early phase of the pilot study, all operated fetuses had to undergo standard postnatal repair of the lesion. Encouraged by the observation that our protocol for materno-fetal anesthesia and amniotic carbon dioxide insufflation were well tolerated by both mothers and fetuses over hours, surgical dissection of the lesion followed by closure with a collagen patch was introduced in subsequent cases (7,20).
At that time, I believed that the collagen patch employed for watertight coverage of the lesion would be overgrown by skin until delivery. I had observed the remarkable wound healing capabilities in fetal sheep, in which large experimental spina bifida-like lesions almost entirely healed within a few weeks after surgery. Yet, in human fetuses that miracle did not happen and the surface of the patch did not overgrow at all or only to some degree when the procedure was performed prior to 22 weeks of gestation. Therefore, in almost all of or our operated fetuses, patch overgrowth with skin still has to
This article is protected by copyright. All rights reserved
happen within the first postnatal weeks of life as most lesions (Fig. 5). Nevertheless, in fetuses with less wide malformations, direct skin closure above the patch is also possible (Fig. 3).
Further advances in handling the fetoscopic setup and increasing experience with the surgical anatomy of the malformation have helped to adapt the fetoscopic approach to lesions of various sizes and surgical complexities: Single-patch closure with a collagen patch is now the most commonly performed technique. It is prefered in those cases, in which the spinal canal is deep enough such that cerebrospinal fluid can reaccumulate between spinal cord and patch, reducing the risk of tethering of the neural cord to the patch in the postoperative period. In more shallow lesions that do not allow for a sufficient cerebrospinal fluid cushion between spinal cord and patch or in cases where – given the limitations of fetoscopy – small skin remnants can sometimes not be completely removed during fetoscopy, double-patch closure is performed: the spinal cord tissue is first covered with a small inert teflon patch, then the entire defect is covered with an additional collagen patch (Fig. 2).This approach is chosen in order to facilitate the conditions for postnatal neurosurgical procedures. In fetuses with giant malformations, I used to cover the operated region with an additional teflon patch that served as a wound dressing and permitted at least some degree of prenatal skin closure underneath it. This step is now rarely performed because in most current procedures the normal skin surrounding the malformation is being pulled over the patch as far as possible. This step largely decreases the patch surface that remains exposed to the amniotic milieu. As described above, in some smaller lesions, even complete skin closure can now be achieved with the fetoscopic approach, facilitating and shortening the length of postnatal wound treatment of these newborns (Fig 3). In these cases in order to obviate the need for postnatal intervention, our neurosurgeon recommended that direct fetal skin closure should be preceeded by coverage of the neural tissue with a patch. This step provides the operated region with more stability. The technically more simple fetoscopic closure possibilities have come under attack from some colleagues performing the open operative approach. Their line of thinking is that during fetal repair, the postnatal standard of repair in three layers must be adherered to (21). This scientifically unfounded notion has been challenged by Pedreira and colleagues from Sao Paulo who themselves have developed a simplified closure technique in sheep (22) which they recently introduced clinically.
This article is protected by copyright. All rights reserved
In a controlled study in sheep, their simplified technique for fetoscopic spina bifida closure resulted in better preservation of nerve tissue and less adherence of the spinal cord to the scar when compared to the classic technique performed by a neurosurgeon. Their findings prompted them to “suggest that the use of the current (classic neurosurgical 3-layer) repair technique (during open fetal surgery) may need to be reassessed“ (22). Clearly, the future will show to what extent the fetal malformation can or must be surgically addressed prenatally. In order to facilitate surgical dissection during more complex procedures (Fig. 6), we and others suggested and tested the incorporation of surgical robotics for minimally-invasive fetal SBA repair in sheep (8,23). The results of these feasibility studies did not provide any evidence for superior patch coverage of experimental skin lesions by a robot compared to what can be achieved by a manual approach. Yet in order to facilitate fetoscopic dissection of the placode and achieve the precision of tissue handling that would be needed for a fetoscopic “classic 3-layer approach“, surgical robotics would be desirable. Once the malformation has been dissected and covered with patch material, a feat that is achieved in about three hours now in most fetuses, carbon dioxide is removed during simultaneous refilling of the amniotic cavity with a warm sterile crystaline solution. The focus at this stage lies on coverage of the trocar defects within the fetal membranes with especially designed uterine closure devices. Yet, the hope that coverage of the membranes would reduce the incidence of amniotic fluid leakage with ongoing gestation has not yet been fulfilled. Regardless whether all trocar insertion sites were covered or not at the time of surgery, transvaginal amniotic fluid loss occurs in most patients at a mean gestational age of 30 weeks. I believe that dislodgment of closure devices or destruction of their fragile membrane from fetal tampering are the main causes for this observation. Nevertheless, coverage of the trocar insertion defects within the chorioamniotic membranes has been one of the most important accomplishments during the development of the fetoscopic technique. Since then nearly 90% of fetuses have been born at or beyond 30 weeks of gestation and benefit from the low rates of serious complications of prematurity nowadays achieved by modern neonatology at this age. In contrast, the earlier reported and heavyly critizised series of fetuses that I had operated during the pilot phase of the approach was delivered at a mean gestational age of 29 weeks (11,24). At the time, special membrane coverage devices had not been developed yet and wedging collagen
This article is protected by copyright. All rights reserved
plugs into the uterine wall during removal of each trocar proved to be as ineffective then as in more recent single trocar procedures (25,26).
This article is protected by copyright. All rights reserved
Conclusion In conclusion, percutaneous minimally-invasive fetoscopic surgery for SBA in human fetuses can be performed with a high rate of technical success regardless of placental position. Despite its potential for important benefits in patients with SBA, there is still a low risk for perioperative fetal demise or significant chronic morbidity from prematurity.
Acknowledgment This paper is dedicated to Claudia and our wonderful boys
This article is protected by copyright. All rights reserved
References 1. Adzick NS, Thom EA, Spong CY, Brock JW, Burrows PK, Johnson MP, Howell LJ, Farrel JA, Dabrowiak ME, Sutton LN, Gupta N, Tulipan NB, D’Alton ME, Farmer DL. A randomized trial of prenatal versus postnatal repair of myelomeningocele. N Engl J Med 2011; 364(11) : 993-1004. 2. Bruner JP, Tulipan NE, Richards WO. Endoscopic coverage of fetal open myelomeningocele in utero. Am J Obstet Gynecol 1997; 176 : 256-257. 3. Bruner JP, Tulipan NB, Richards WO, Walsh WF, Boehm FH, Vrabcak EK. In utero repair of myelomeningocele: a comparison of endoscopy and hysterotomy. Fetal Diagn Ther 2000; 15 : 83-88. 4. Farmer DL, von Koch CS, Warwick JP, Danielpour M, Gupta N, Lee H, Harrison MR. In utero repair of myelomeningocele. Arch Surg 2003; 138 : 872-878. 5. Kohl T, Witteler R, Strümper D, Gogarten W, Asfour B, Reckers J, Merschhoff G, Marcus AE, Weyand M, Van Aken H, Vogt J, Scheld HH. Operative techniques and strategies for minimally invasive fetoscopic fetal cardiac interventions in sheep. Surg Endosc 2000; 14 : 424-430. 6. Kohl T, Szabo Z, Suda K, Harrison MR, Quinn TM, Petrossian E, Hanley FM. Percutaneous fetal access and uterine closure for fetoscopic surgery - Lessons from 16 consecutive procedures in pregnant sheep. Surgical Endoscopy 1997; 11 : 819-824. 7. Kohl T, Tchatcheva K, Weinbach J, Hering R, Kozlowski P, Stressig R, Gembruch U. Partial amniotic carbon dioxide insufflation (PACI) during minimally invasive fetoscopic surgery – Early clinical experience. Surg Endosc. 2010; 24 : 432-444. 8. Kohl T, Große Hartlage M, Kienitz D, Westphal M, Buller T, Aryee S, Achenbach S, Gembruch U, Brentrup A. Percutaneous fetoscopic patch coverage of experimental lumbosacral full- thickness skin lesions in sheep - A minimally invasive technique aimed at minimizing maternal trauma from fetal surgery for myelomeningocele. Surg Endoscopy 2003; 17 : 1218-1223. 9. Kohl T, Hering R, Heep A, Schaller C, Meyer B, Greive C, Bizjak G, Buller T, Van de Vondel P, Gogarten W, Bartmann P, Knöpfle G, Gembruch U. Percutaneous fetoscopic patch coverage of spina bifida aperta in the human – Early clinical experience and potential. Fetal Diagn Ther 2006; 21 : 185193. 10. Kohl T, Tchatcheva K, Merz W, Wartenberg HC, Heep A, Müller A, Franz A, Stressig R, Willinek W, Gembruch U. Percutaneous fetoscopic patch closure of spina bifida aperta in the human –
This article is protected by copyright. All rights reserved
Advances in fetal surgical techniques may now obviate the need for early postnatal neurosurgical intervention. Surg Endosc 2009; 23 : 890-895. 11. Verbeek RJ, Heep A, Maurits NM, Cremer R, Hoving EW, Brouwer OF, van der Hoeven JH, Sival DA. Fetal endoscopic myelomeningocele closure preserves segmental neurological function. Dev Med Child Neurol. 2012; 54 : 15-22. 12. Kohl T. Stool contamination. Neurosurg Pediatrics 2010; 5 : 422. 13. Kohl T, Kawecki A, Degenhardt J, Axt-Fliedner R. Preoperative sonoanatomic examination of fetal spina bifida permits prediction of surgical complexity during subsequent minimally-invasive fetoscopic closure. Ultrasound Obstet Gynecol 2012; 40 : 8. 14. Kohl T, Hering R, Van de Vondel P, Tchatcheva K, Berg C, Bartmann P, Heep A, Franz A, Müller A, Gembruch U. Analysis of the step-wise clinical introduction of experimental percutaneous fetoscopic surgical techniques for upcoming minimally-invasive fetal cardiac interventions. Surgical Endosc 2006; 20 : 1134-1143. 15. Kohl T, Ziemann M, Weinbach J, Tchatcheva K, Hasselblatt M. Partial amniotic carbon dioxide insufflation (PACI) during minimally invasive fetoscopic interventions seems safe for the fetal brain in sheep. J Laparoendosc Surg 2010; 20 : 651-653. 16. Kohl T, Reckers J, Strümper D, Grosse Hartlage M, Gogarten W, Gembruch U, Vogt J, Van Aken H, Scheld HH, Paulus W, Rickert CH. Amniotic air insufflation during minimally invasive fetoscopic fetal cardiac interventions is safe for the fetal brain in sheep. J Thorac Cardiovasc Surg. 2004 ; 128 : 467-471. 17. Luks FI, Deprest JA, Marcus M, Vandenberghe K, Vertommen JD, Lerut T. Carbon dioxide pneumamniosis causes acidosis in fetal lambs. Fetal Diagn Ther 1994; 9 : 105-109. 18. Gratacos E, Wu J, Devlieger R, Van de Velde M, Deprest JA. Effects of amniodistension with carbon dioxide on fetal acid-base status during fetoscopic surgery in the sheep model. Surg Endosc 2001; 15 : 368-372. 19. Sayki Y, Litwin DE, Bigras JL, Waddel J, Konig A, Baik S, Navsarikar A, Rebeyka IM. Reducing the deleterious effects of intrauterine CO2 during fetoscopic surgery. J Surg Res 1997; 69 : 51-54. 20. Hering R, Hoeft A, Putensen C, Tchatcheva K, Stressig R, Gembruch U, Kohl T. Maternal haemodynamics and lung water content during percutaneous fetoscopic interventions under general anaesthesia. Brit J Anaesth 2009; 102 : 523-527.
This article is protected by copyright. All rights reserved
21. Meuli M, Moehrlen U. Fetal surgery for myelomeningocele: a critical appraisal. Eur J Pediatr Surg. 2013; 23 : 103-109. 22. Herrera SR, Leme RJ, Valente PR, Caldini EG, Saldiva PH, Pedreira DA. Comparison between two surgical techniques for prenatal correction of meningomyelocele in sheep. Einstein (Sao Paulo). 2012; 10 : 455-461. 23. Aaronson OS, Tulipan, NB, Cywes R, Sundell HW, Davis GH, Bruner JP, Richards WO. Robotassisted endoscopic intrauterine myelomeningocele repair: a feasibility study. Pediatr Neurosurg 2002; 36 : 85-89. 24. Shurtleff D. Fetal endoscopic myelomeningocele repair. Dev Med Child Neurol. 2012; 54 : 4-5. 25. Kohl T. Iatrogenic fetal membrane damage from complex fetoscopic surgery in human fetuses might not be amenable to simple closure by collagen plugs. Prenat Diagn 2008; 28 : 876-877. 26. Engels AC, Van Calster B, Richter J, Dekoninck P, Lewi L, De Catte L, Devlieger R, Deprest JA. Collagen plug sealing of iatrogenic fetal membrane defects after fetoscopic surgery for congenital diaphragmatic hernia. Ultrasound Obstet Gynecol 2013. DOI: 10.1002/uog.12547.
This article is protected by copyright. All rights reserved
ID Age Lesion level
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
31 30 23 31 36 36 22 34 29 38 35 30 19 36 30 30 30 19 41 40 38 36 29 34 40 25 28 28 33 36
Placental position Gestational age at fetal surgery Gestational age at delivery ( Anterior / Posterior) skin-to-skin-time weeks+days weeks+days P 24+0 34+2 230 A 22+6 29+4 240 A 25+0 34+0 240 A 24+0 36+5 255 A 23+4 24+4 270 P 21+1 34+0 240 A 25+0 30+4 290 P 24+6 30+0 300 A 25+3 30+4 260 A 21+5 32+3 195 A 25+3 30+5 265 P 27+2 36+4 270 P 22+0 34+3 180 P 23+3 34+1 210 P 25+1 32+0 315 P 24+4 38+1 255 P 22+1 34+3 270 P 24+1 31+0 240 P 22+6 30+0 300 A 22+4 34+3 210 A 24+6 33+4 210 A 21+4 28+5 215 P 23+0 30+0 210 A 21+3 33+0 190 P 23+3 37+2 210 P 25+4 34+1 225 P 22+2 34+6 225 P 22+4 37+2 225 P 24+4 31+3 245 P 23+1 29+6 245
This article is protected by copyright. All rights reserved
L4 L1 L5 L4 L3 L2 L2 L5 L5 L4 L3 S1 L4 L4 L4 L5 L1 L4 L1 L4 L5 L3 L4 L3 L5 L3 L4 L4 L3 L5
Lesion type flat / cystic f f f f f f f f c f f f c c c f f f f f f f f f f c f c f f
Lesionsize (cm) Amniotic access achiev 3 4x3 4x3 4x3 4x3 4x3 4x2 3 2 2 x 1.5 3.5 x 3 3x2 3 3 3 2 x 1.5 4x3 3 x 1.5 3 x 3.5 2.5 x 3 3x2 2 x 1.5 2 2.5 2.5 4.5 x 3 4 x 2.5 4 3 x 2.5 3 x 2.5
+ + + + + + + + + + + + + + + + + + + + + + + + + + + + + +
31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51
34 23 30 40 33 29 26 23 37 25 26 32 32 30 38 37 32 28 32 30 28
P A A A P P P P A A P A P P P P P P A P A
24+0 24+4 23+6 21+0 26+6 21+4 22+4 22+6 24+3 23+1 28+5 22+5 24+0 29+1 22+3 23+0 23+2 23+6 22+4 23+1 23+4
29+2 32+5 35+1 31+1 38+1 34+2 33+6 32+2 30+2 34+2 34+0 32+1 30+6 36+0 34+2 34+4 31+0 34+1 27+6 35+1 34+1
205 45 240 165 160 150 155 200 225 180 140 220 200 210 185 190 190 225 280 175 225
L1 L3 L5 L4 L5 L5 L5 T 11 L5 L2 L5 L5 Th 11 L4 L4 L3 L3 S1 L2 L5 L2
f f f c f c f f c f f f f f c c c f c c c
3 x 2.5 n.a. 3 x 2.5 3 x 2.5 3x2 2 3 x 2.5 4x3 4x3 5x3 4 x 3.5 3.5 cm 5x4 4x3 5x4 5x4 3.5 x 3 3.5 x 3 4 x 3.5 3x2 3.5 x 3
- n.a. = not applicable; DC = direct closure; DCP = direct closure over patch; DP = double patch closure; SP = single patch closure; TP = triple patch closure
This article is protected by copyright. All rights reserved
+ + + + + + + + + + + + + + + + + + + + +
Legends
Figure 1: External aspect of the percutaneous setup - Percutaneous minimally-invasive fetoscopic surgery for SBA is performed via three (to four) trocars with an external diameter of 5 mm that are placed into the amniotic cavity under continuous ultrasound monitoring by a Seldinger approach (top). Following partial amniotic carbon dioxide insufflation (PACI) and fetal posturing, surgical dissection of the malformation is carried out using fetoscopic instruments (bottom).
This article is protected by copyright. All rights reserved
Figure 2: Minimally-invasive fetoscopic double patch closure of a large L3-myelomeningocele at 23+2 weeks of gestation. Top row – left – The large myelomeningocele before surgical dissection. The exposed neural tissue can be visualized clearly in the gas-insufflated amniotic cavity. Top row right - The malformation is being circumcised with a needle electrode. Middle row – left - Note the high degree of precision that can be achieved within less than a millimeter range during manual dissection of the placode using micro-grasper and micro-scissors. Middle row – right: Fetoscopic view of the neural tissue after removal of surrounding pathological tissues. Note the excellent visualization of the spinal cord nerves in the high-powered fetoscopic field employing a 3.3 mm-30° rod-lens fetoscope. Bottom row – left: Fetoscopic view after protecting the neural tissue with an inert Gore⎝-PTFE patch. This step aims at the prevention of tethering. Bottom row – right: Definitive coverage of the entire surgical field is then achieved with a collagen patch. The goal of watertight coverage is demonstrated at the end of the procedure by bulging of the patch as well as lack of cerebrospinal fluid leakage when compressed with an instrument.
This article is protected by copyright. All rights reserved
Figure 3: Minimally-invasive fetoscopic direct closure of a sacral myeloschisis at 23+6 weeks of gestation. Top row – left – The sacral myeloschisis before surgical dissection. A small amount of stool (arrow) can be visualized over the opening of the central canal at the rostral end of the neural tissue. Top row - right – Following surgical dissection, the malformation has been covered with a collagen patch. Bottom row – left – Then, complete skin closure has been achieved. Bottom row – right: After delivery, only a very small part of the collagen patch can be seen as a result of dehissence of one of the skin sutures. The skin overgrew the patch without the need for any further intervention. Figure 4: Top - Employing ultrasound-guidance, percutaneous fetal access with three trocars can be achieved in essentially all cases. As each trocar is inserted over a wire guide, the intial puncture hole for maternal percutaneous-transabdominal-transuterine-intraamniotic trocar insertion has a diameter of only 1.2 mm. Bottom – The tiny incisions and punctures for trocar insertion during the minimally-invasive approach are barely comparable to the much larger incisions in the maternal abdomen and uterus required for open fetal surgery.
This article is protected by copyright. All rights reserved
This article is protected by copyright. All rights reserved
Figure 5: Top – Newborn with spina bifida at 35 weeks of gestation following double-patch closure of an L4 defect extending over the entire sacrum at 22+4 weeks of gestation. As in human fetuses the surface of the patch does not overgrow with skin over the remainder of gestation, this step has to happen within the first postnatal weeks of life. Middle – The perfectly ingrown upper patch after removal of the surgical clips. Bottom – Fetoscopic image of the large malformation with a long segment of exposed placode (arrow) before closure. Note that at the time of surgery, many malformation are almost as wide as the entire fetal back.
This article is protected by copyright. All rights reserved
This article is protected by copyright. All rights reserved
Figure 6: Surgically challenging setups for minimally-invasive fetoscopic surgery of spina bifida that may benefit from the incorporation of surgical robotics. Top left – Large L4-myelomeningocele with a terminal myelocystocele (m) at 25+1 weeks of gestation. The main surgical difficulty resulted from sparing the spinal nerves (arrows) that were attached along the inner surface of the sac from injury during resection of the sac tissue. Top right – after removal of most parts of the sac, the intact nerves (arrows) can be seen still adherent to a part of the inner layer of the sac that was preserved. Middle left - Large L5-myelomeningocele. The delicate nerves of the cauda equina (ce) are found attached (arrow) to the far sturdier pathological skin at 24+3 weeks of gestation. Middle right – Fetoscopic image after removal of all sac tissue. Bottom left – L5-myelomeningocele with pathologic skin adjacent to the neural placode (p) at 26+6 weeks of gestation. Beige-colored stool can be seen along the placode and also on its surface. As in the case depicted in the middle row, the surgical difficulty here results from the dissection of the delicate neural tissue from the far sturdier pathological skin. Bottom right – In this case, a firm strand of amniotic tissue divides the amniotic cavity in two compartments. This problem makes posturing maneuvers more difficult and in itself poses a risk factor for preterm rupture of membranes.
This article is protected by copyright. All rights reserved
This article is protected by copyright. All rights reserved
Corresponding author: Thomas Kohl MD German Center for Fetal Surgery & Minimally Invasive Therapy (DZFT) University Hospital Giessen-Marburg Klinikstr. 33 35592 Giessen – Germany
phone: +49-175-597-1213 fax: +49-641-985-45279 [email protected] www.dzft.de
Keywords fetoscopy, fetal surgery, spina bifida, Chiari-II-malformation, hydrocephalus, fetus
This article has been accepted for publication and undergone full peer review but has not been through the copyediting, typesetting, pagination and proofreading process, which may lead to differences between this version and the Version of Record. Please cite this article as doi: 10.1002/uog.13430
This article is protected by copyright. All rights reserved
Abstract Background Percutaneous minimally-invasive fetoscopic closure of spina bifida aperta (SBA) was developed by the author in order to obviate the need for the far more invasive open fetal surgical approach. The manuscript offers an analysis of the current technical approach accompanied by an orienting overview on its development in ovine and human fetuses. Flanked by supporting manuscripts about management and outcome, the author hopes to raise the interest in the further dissemination of minimally-invasive surgery for this condition. Patients & Methods Minimally-invasive percutaneous fetoscopic closure of SBA was performed at the German Center for Fetal Surgery & Minimally-Invasive Therapy (DZFT) in 51 human fetuses between 21.0 and 29.1 weeks of gestation (mean age 23.8 weeks). Various parameters of surgical relevance for the success and safety of the procedure and the early perioperative outcome were retrospectively analyzed. In addition, historical information from the early clinical cases and how it shaped the mature approach is is being discussed. Results Percutaneous minimally-invasive fetoscopic closure of SBA was performed with a high rate of technical success, regardless of placental or fetal position. Fetal demise one week after the procedure occurred in one of the 51 cases (2%) from early postoperative chorioamnionitis. All other fetuses survived gestation. Of the surviving fetuses 45 (90%) were born beyond 30 weeks and 25 (49%) beyond 34 weeks of gestation. Conclusions Following an adequate learning curve, minimally-invasive fetoscopic surgery for fetal spina bifida be performed with a high rate of technical success regardless of placental position.
This article is protected by copyright. All rights reserved
Introduction Open fetal surgery for spina bifida aperta (SBA) has been performed in hundreds of human fetuses. The Management-Of-Myelomeningocele-Study (MOMS) provided level I - evidence that, compared to prenatally unoperated fetuses, prenatal SBA closure may preserve leg function and reduce the severity of hindbrain herniation and hydrocephalus in affected fetuses (1). Yet, the open approach is associated with significant maternal morbidity as it requires maternal laparotomy and hysterotomy for fetal access. In a significant proportion of cases, the hysterotomy scar left after open fetal surgery becomes a weak spot, prone to uterine wall dehissence or even rupture after fetal surgery or in future pregnancies.
As the clinical consequences of SBA provide multiple reasons for life-long disabilities, sorrows and requirements for therapy, the maternal trauma from the open approach has been accepted by many expectant mothers and prenatal specialists as the prize to be paid for the clinical benefits. Yet, even the earliest prenatal procedures in the mid-1990s already aimed at reducing maternal injury: In order to avoid maternal hysterotomy, Drs. Bruner and Tulipan performed their pioneering surgeries by fetoscopy, securing skin grafts to the malformation (2,3). Their technique, however, still required maternal laparotomy followed by transuterine trocar placement. It was quickly abandoned in favor of the open approach because of technical difficulties and unfavorable outcomes. A few other attempts at fetoscopy were performed at UCSF but subsequently also abandoned for similar reasons (4).
During studies in inanimate models, sheep and postmortem human fetuses, the author developed an even less invasive because fully percutaneous minimally-invasive fetoscopic approach for prenatal closure of SBA (5-10). This manuscript offers an orienting overview of its development and presents technical data as well as perioperative outcomes from 51 cases that were performed at the German Center for Fetal Surgery & Minimally-Invasive Therapy (DZFT) over the past three years. Flanked by accompanying manuscripts about selection, perioperative management and the mostly promising neonatal and 30-month neurological outcomes, the author wishes to promote the dissemination and further development of minimally-invasive fetoscopic surgery for SBA.
This article is protected by copyright. All rights reserved
Patients & Methods Various parameters of surgical relevance for the success of percutaneous minimally-invasive fetoscopic spina bifida closure were retrospectively analyzed in 51 cases that were performed at our center between July 2010 to June 2013 (Table). Maternal age at surgery ranged between 19 and 41 years with a mean age of 31.2 years. Gestational age at surgery ranged between 21.0 and 29.1 weeks of gestation with a mean age of 23.8 weeks.
Analysis of this data has been approved by the local committee on human research (#158/13). The procedure has been offered as standard of care at the German Center for Fetal Surgery & MinimallyInvasive Therapy (DZFT) since July 2010 following 1. completion of a CHR-approved pilot study in 30 patients at the University of Bonn, Germany, 2. when an independent match-controlled study in 13 of these pilot study patients showed that postnatal leg function was significantly better and the rate of shunt-dependant hydrocephalus was significantly lower than in prenatally unoperated controls (11), and 3. when the technical success rate, as well as maternal and fetal safety, and gestational age at delivery had been greatly improved upon compared to the early pilot phase experience.
Fetal selection The gestational age of the fetuses undergoing percutaneous fetoscopic patch coverage of SBA preferably ranges between 20+0 and 25+6 days. Only a few procedures are carried out later when the degree of hydrocephalus is still mild and stool soiling of the lesion can be anticipated from characterization of ist surgical anatomy by ultrasound (12,13). The defect should be located between Th1 to S1. Fetuses with a gibbus are excluded. Further requirements are hindbrain herniation and a diameter of the lateral brain ventricles of less than 16 mm as assessed by ultrasound and MRI. The fetuses should be free from other major organ abnormalities on ultrasound and present a normal karyotype. Yet some expectant women, following counseling, elect to forgo preoperative amniocentesis. In these cases, fetal surgery is offered when ultrasound- and MRI-studies do not raise
This article is protected by copyright. All rights reserved
suspicions of any other major fetal malformations or chromosomal anomalies. Following this selection policy, one fetus, whose parents had opted against preoperative karyotyping, that underwent technically successful surgery and was born at 36 weeks of gestation, died from trisomy 13 after delivery. This case was excluded from this analysis.
Intraoperative ultrasound-monitoring Apart from providing the fetal patient in the first place, multimodal ultrasound provides information critical for the safety and success of minimally-invasive fetoscopic closure of SBA: Surgically relevant information about the lesion that helps in planning the procedure, selecting safe and suitable trocar sites during intraamniotic access, monitoring of uterine closure and last but not least hemodynamic information is collected. Doppler ultrasound interrogation of the fetoplacental circulation as well as determination of fetal heart rate and contractility are regularly performed at defined times: before and after induction of anesthesia, after providing the mother with central venous and arterial lines, before intraamniotic access, after amnioinfusion, before insufflation and at the end of the procedure. This hemodynamic information helps the anesthesiologist to adjust the depth of anesthesia and to optimize maternal and fetal hemodynamics by maternal administration of volume, catecholamines and other drugs to the requirements of the procedure in dosages that warrant safety of both. Only when maternal and fetal hemodynamics remain stable after induction of anesthesia and, if required for achieving sufficient intraamniotic work space, amnioinfusion, the first trocar will be placed. If maternal or hemodynamics or fetoplacental blood flow deteriorate after induction of anesthesia or amnioninfusion, and attempts at their normalization are in vain, a procedure will be abandoned. Fortunately, this situation is rare and has been encountered only once over the past three years.
Surgical approach Percutaneous minimally-invasive fetoscopic surgery for SBA is performed via three (to four) trocars with an external diameter of 5 mm that are placed into the amniotic cavity under continuous ultrasound monitoring by a Seldinger approach (Fig 1). Following partial removal of the amniotic fluid, partial amniotic carbon dioxide insufflation (PACI) is performed as previously described in order to
This article is protected by copyright. All rights reserved
maintain visibility of intraamniotic contents (4,6). Fetal posturing is then carried out using fetoscopic instruments (Karl Storz, Tuttlingen, Germany). Then, the lesion is being circumcised with a needle electrode and the neural tissue is carefully freed from surrounding tissues (Fig 2 & 3). Depending on the anatomy of the lesion, the neural cord is covered with one or more patches. Regardless of the closure technique, the goal of watertight coverage is demonstrated at the end of the procedure by observing bulging of the patch as well as lack of cerebrospinal fluid leakage when its surface is being compressed with an instrument. Once the lesion has been closed, the insufflated carbon dioxide is being removed during simultaneous refilling of the amniotic cavity with a warm sterile crystaline solution. Following percutaneous coverage of the membranous defects at the trocar insertion sites, the abdominal trocar insertion sites are closed with non-absorbable 5-0 surgical sutures.
Study variables The following parameters were analyzed: Gestational age at delivery, acute mortality of surgery, placental position, success of percutaneous fetal access, number of trocars, incidence and consequences of trocar dislodgment, success of fetal posturing, pressures and duration of partial amniotic carbondioxide insufflation (PACI), need for additional venting of insufflation gas from the maternal abdomen, type of lesion (flat, cystic), lesion size, fetoscopic handling of the lesion (patch closure, direct closure, combination of both), overall technical success of the procedure defined as patch coverage of the neural tissue, success of uterine closure, and skin-to-skin-time of fetoscopic procedure.
Results Gestational age at delivery in survivors ranged between 27.9 and 38.1 weeks of gestation with a mean age of 33.0 weeks. All fetuses survived surgery and all but one gestation. The fetus that died, died from preterm delivery due to chorioamnionitis at 24.6 weeks about one week after the procedure (Case 5; Table). During the first months of life, two more infants died from severe brain-stem dysfunction attributed to the Chiari II – malformation (Cases 3 & 7; Table).
This article is protected by copyright. All rights reserved
20 procedures were performed in the presence of an anterior and 31 procedures in the presence of a posterior placenta. Skin-to-skin time ranged between 45 and 315 minutes. The 45-minute-case was abandoned because maternal obesity in combination with an anterior placenta made completion of the procedure technically impossible. The shortest successful case took 140 minutes in a fetus with a flat lesion that could be easily dissected and covered by a single-patch. Thirteen procedures were performed between 140 to 200 minutes, 25 procedures took between 201 – 250 minutes. Eleven procedures were performed between 251 and 301 minutes. The longest case took 315 minutes. In this case, the surgical anatomy was extremely complex and triple patch coverage had been performed. Ultrasound-guided percutaneous intraamniotic access with ports with an external diameter of 5 mm was established in all cases. During 49 procedures three and during two procedures four trocars (Cases 5 & 7; Table) were inserted into the amniotic cavity. In cases 5 and 7, the additional trocar was required for maintaining a stable fetal position throughout the procedure. In one case, dislodgment of one trocar occured at the end of one procedure such that closure of the trocar insertion site could not be performed (Case 24; Table). In another case that had to be abandoned, dislodgment of two trocars occurred right after induction of insufflation as a consequence of maternal obesity (Case 32; Table). Favorable percutaneous fetal posturing was achieved in all but the one case where early dislodgment of trocars occurred. Opening pressures for PACI ranged between 9 to 30 mm Hg (mean 15.6 mm Hg) and duration ranged between 10 – 275 minutes (mean 183 min; Table). PACI was well tolerated by both mother and fetus and permitted clear visualization throughout the entire fetoscopic procedure. Additional venting of insufflation gas from the maternal peritoneal cavity was required in 17 cases. There were 35 flat and 15 cystic lesions. Measured at skin level, their areas ranged between 3 to 12 cm2. Overall technical success of the procedure defined as complete patch coverage of the exposed neural tissue at the end of fetal surgery was achieved in all but the one case where early dislodgment of two trocars occurred in an obese patient (Case 32; Table). Single-patch covererage was performed in 30 fetuses, double-patch coverage in 14 (Fig. 2), triple-patch coverage in two and direct closure of the lesion in four fetuses. In two fetuses of the latter group, the skin was closed immediately above the neural cord without interposition of patch material (Cases 35 & 44; Table). In the two other ones,
This article is protected by copyright. All rights reserved
and recommended by our local neurosurgeon, direct skin closure was performed after covering the neural cord with a collagen patch (Cases 48 & 51; Table; Fig. 3). Coverage of all membranous defects at trocar removal was achieved in 47 cases, coverage of only two of three access sites was achieved in the remaining 4 cases. The reasons for not achieving or forgoing coverage were trocar dislodgment in two cases, avoiding the risk of umbilical cord compromise in one case and two too closely situated trocar insertion sites in another one.
This article is protected by copyright. All rights reserved
Discussion This retrospective analysis shows that percutaneous minimally-invasive fetoscopic surgery for SBA in human fetuses can be performed with a high rate of technical success regardless of placental position, and with a low rate of perioperative mortality. Like the open fetal surgical approach, percutaneous fetoscopic patch coverage of the malformation aims at protecting the spinal cord tissue, reversing hindbrain herniation and reducing the need for postnatal ventriculo-peritoneal shunt insertion. Compared to open fetal surgery for SBA, however, the percutaneous minimally-invasive fetoscopic approach offers a major reduction of maternal injury by avoiding laparotomy and hysterotomy. Hence, maternal pain and discomfort are usually minimal beyond the 2nd day after surgery. In addition, most expectant women can be discharged home from hospital within a week. Adhering to the fundamental ethical principle of primum non nocere, further exploration and dissemination of the minimally-invasive fetoscopic techniques seem worthy goals and may over time allow replacing the open operative approach. Employing ultrasound-guidance, percutaneous fetal access with three trocars can be achieved in essentially all cases. Rarely, a fourth trocar is required for securing a favorable fetal lie throughout the procedure. As each trocar is inserted over a wire guide, the intial puncture hole for maternal percutaneous-transabdominal-transuterine-intraamniotic trocar insertion has a diameter of only 1.2 mm. This gentle method of intraamniotic access is the hallmark of percutaneous minimally-invasive fetal surgery and is barely comparable to the much larger incisions in the maternal abdomen and uterus required for open fetal surgery (Fig. 4). Whereas I assumed in the beginning that percutaneous minimally-invasive fetoscopic spina bifida closure might only be feasible in women with posterior, lateral or fundal placentae, growing clinical experience showed that the operation can be performed in most cases regardless of placental position. During ultrasound-guided trocar insertion in cases with an anterior placenta, a safety margin of about 3 cm away from the placental edge suffices in preventing placental injury. Trocar dislodgement resulting in loss of surgical setup has been fortunately rare in this clinical series. It usually suffices to advance each trocar for about 3-5 cm into the amniotic cavity. Yet, as seen in two of our cases, trocar dislodgment may occur during instrument manipulations or in adipose patients in whom the trocar tip can sometimes be advanced for only 2 cm or less into the amniotic cavity. Trocar dislodgment can also be the consequence of too much traction between the abdominal
This article is protected by copyright. All rights reserved
and uterine walls or of sudden uterine volume changes from intraoperative uterine contractions or when the amniotic cavity gradually collapses from unattended gas loss along the trocars into the maternal abdomen (5). In cases with an anterior placenta, trocar dislodgment requires to abandon the procedure since amniotic gas insufflation precludes ultrasound-guided reinsertion of another trocar. Unmonitored or in the presence of impaired ultrasound imaging quality, trocar insertion in cases with an anterior placenta carries a high risk for placental or maternal bowel injury. In contrast, in cases with a posterior placenta, reinsertion can easily be achieved through the original opening guided by fetoscopy. Successful fetal posturing with endoscopic instruments can usually be achieved in most cases. Yet, the more obese a patient, the more difficult posturing maneuvers can become. In this situation, the fragile instruments and endoscopes oftentimes bend to critical degrees. In obese patients with a posterior placenta postponing the procedure until the fetus lies in a more favorable position has been a helpful strategy. Alternatively, when more severe obesity comes with an anterior placenta, maternal laparotomy and transuterine trocar placement or even open fetal surgery may be considered. Due to the small trocar size, continuous pressure- and volume-controlled exchange of large volumes of crystalline solutions in order to manage bleeding complications or provide a clear field for dissecting and closing the malformation throughout the entire procedure may be difficult or impossible to achieve. In order to overcome the disadvantages of operating in fluid, management principles and safety measures for percutaneous gas insufflation of the amniotic cavity were defined from animal studies in sheep and early clinical experience (5,6,7,14,15). PACI over a mean period of three hours permitted clear visualization throughout all 51 fetoscopic procedures and did not result in any critical or unmanageable maternal cardiovascular effects as assessed by arterial blood gas analyses, continuous maternal ECG, blood pressure, oxygen saturation and end-expiratory carbon dioxide concentration monitoring. Encouragingly, postnatal clinical examinations as well as ultrasound- and MRI-scanning in the infants did not reveal any evidence of chronic brain injury that could be attributed to the fetoscopic setup. These clinical findings were supported by previous insufflation studies of our group in sheep (15,16). During the earliest clinical cases, it was entirely unclear how well and for how long mother and fetus would tolerate anesthesia and in particular PACI. At this time, my attempts at securing funding for amniotic insufflation studies in primates with the goal of studying various issues relevant to maternal
This article is protected by copyright. All rights reserved
safety of the novel technique were in vain. Years earlier, Bruner et al. were the first who insufflated an exteriorized uterus in humans during four fetoscopic procedures and encountered placental abruption in one of them (2). In contrast to their approach and making matters more complicated, PACI had never been used before in a percutaneous fetoscopic setup in humans. Further discouraging, previous studies of other investigators on exteriorized sheep suggested that carbon dioxide and/or insufflation pressures higher than 15 mm Hg would lead to dangerous degrees of fetal acidosis (1719). During own studies in sheep (a species that has a syndesmochorial placenta), I observed that insufflation of the amniotic cavity was tolerated at much higher insufflation pressures than previously thought without interfering with fetoplacental blood flow when the maternal abdomen was closed. Nevertheless, the first human procedures in this uncharted territory were kept as short and as simple as possible. Whenever a minor bleeding complication or some membrane detachement was detected, the procedure was abandoned for maternal safety considerations. As a result of that caution, three of the first 19 procedures went uncompleted (11).
In order to keep the procedure short in the initial cases, I simply covered the lesion using inert patch material and left the malformation itself untouched (9). Due to the strong traction forces on the patch rim from fetal growth, the patches often partially dehissed and coverage of the entire lesion until delivery was seldom achieved. As a consequence, during this early phase of the pilot study, all operated fetuses had to undergo standard postnatal repair of the lesion. Encouraged by the observation that our protocol for materno-fetal anesthesia and amniotic carbon dioxide insufflation were well tolerated by both mothers and fetuses over hours, surgical dissection of the lesion followed by closure with a collagen patch was introduced in subsequent cases (7,20).
At that time, I believed that the collagen patch employed for watertight coverage of the lesion would be overgrown by skin until delivery. I had observed the remarkable wound healing capabilities in fetal sheep, in which large experimental spina bifida-like lesions almost entirely healed within a few weeks after surgery. Yet, in human fetuses that miracle did not happen and the surface of the patch did not overgrow at all or only to some degree when the procedure was performed prior to 22 weeks of gestation. Therefore, in almost all of or our operated fetuses, patch overgrowth with skin still has to
This article is protected by copyright. All rights reserved
happen within the first postnatal weeks of life as most lesions (Fig. 5). Nevertheless, in fetuses with less wide malformations, direct skin closure above the patch is also possible (Fig. 3).
Further advances in handling the fetoscopic setup and increasing experience with the surgical anatomy of the malformation have helped to adapt the fetoscopic approach to lesions of various sizes and surgical complexities: Single-patch closure with a collagen patch is now the most commonly performed technique. It is prefered in those cases, in which the spinal canal is deep enough such that cerebrospinal fluid can reaccumulate between spinal cord and patch, reducing the risk of tethering of the neural cord to the patch in the postoperative period. In more shallow lesions that do not allow for a sufficient cerebrospinal fluid cushion between spinal cord and patch or in cases where – given the limitations of fetoscopy – small skin remnants can sometimes not be completely removed during fetoscopy, double-patch closure is performed: the spinal cord tissue is first covered with a small inert teflon patch, then the entire defect is covered with an additional collagen patch (Fig. 2).This approach is chosen in order to facilitate the conditions for postnatal neurosurgical procedures. In fetuses with giant malformations, I used to cover the operated region with an additional teflon patch that served as a wound dressing and permitted at least some degree of prenatal skin closure underneath it. This step is now rarely performed because in most current procedures the normal skin surrounding the malformation is being pulled over the patch as far as possible. This step largely decreases the patch surface that remains exposed to the amniotic milieu. As described above, in some smaller lesions, even complete skin closure can now be achieved with the fetoscopic approach, facilitating and shortening the length of postnatal wound treatment of these newborns (Fig 3). In these cases in order to obviate the need for postnatal intervention, our neurosurgeon recommended that direct fetal skin closure should be preceeded by coverage of the neural tissue with a patch. This step provides the operated region with more stability. The technically more simple fetoscopic closure possibilities have come under attack from some colleagues performing the open operative approach. Their line of thinking is that during fetal repair, the postnatal standard of repair in three layers must be adherered to (21). This scientifically unfounded notion has been challenged by Pedreira and colleagues from Sao Paulo who themselves have developed a simplified closure technique in sheep (22) which they recently introduced clinically.
This article is protected by copyright. All rights reserved
In a controlled study in sheep, their simplified technique for fetoscopic spina bifida closure resulted in better preservation of nerve tissue and less adherence of the spinal cord to the scar when compared to the classic technique performed by a neurosurgeon. Their findings prompted them to “suggest that the use of the current (classic neurosurgical 3-layer) repair technique (during open fetal surgery) may need to be reassessed“ (22). Clearly, the future will show to what extent the fetal malformation can or must be surgically addressed prenatally. In order to facilitate surgical dissection during more complex procedures (Fig. 6), we and others suggested and tested the incorporation of surgical robotics for minimally-invasive fetal SBA repair in sheep (8,23). The results of these feasibility studies did not provide any evidence for superior patch coverage of experimental skin lesions by a robot compared to what can be achieved by a manual approach. Yet in order to facilitate fetoscopic dissection of the placode and achieve the precision of tissue handling that would be needed for a fetoscopic “classic 3-layer approach“, surgical robotics would be desirable. Once the malformation has been dissected and covered with patch material, a feat that is achieved in about three hours now in most fetuses, carbon dioxide is removed during simultaneous refilling of the amniotic cavity with a warm sterile crystaline solution. The focus at this stage lies on coverage of the trocar defects within the fetal membranes with especially designed uterine closure devices. Yet, the hope that coverage of the membranes would reduce the incidence of amniotic fluid leakage with ongoing gestation has not yet been fulfilled. Regardless whether all trocar insertion sites were covered or not at the time of surgery, transvaginal amniotic fluid loss occurs in most patients at a mean gestational age of 30 weeks. I believe that dislodgment of closure devices or destruction of their fragile membrane from fetal tampering are the main causes for this observation. Nevertheless, coverage of the trocar insertion defects within the chorioamniotic membranes has been one of the most important accomplishments during the development of the fetoscopic technique. Since then nearly 90% of fetuses have been born at or beyond 30 weeks of gestation and benefit from the low rates of serious complications of prematurity nowadays achieved by modern neonatology at this age. In contrast, the earlier reported and heavyly critizised series of fetuses that I had operated during the pilot phase of the approach was delivered at a mean gestational age of 29 weeks (11,24). At the time, special membrane coverage devices had not been developed yet and wedging collagen
This article is protected by copyright. All rights reserved
plugs into the uterine wall during removal of each trocar proved to be as ineffective then as in more recent single trocar procedures (25,26).
This article is protected by copyright. All rights reserved
Conclusion In conclusion, percutaneous minimally-invasive fetoscopic surgery for SBA in human fetuses can be performed with a high rate of technical success regardless of placental position. Despite its potential for important benefits in patients with SBA, there is still a low risk for perioperative fetal demise or significant chronic morbidity from prematurity.
Acknowledgment This paper is dedicated to Claudia and our wonderful boys
This article is protected by copyright. All rights reserved
References 1. Adzick NS, Thom EA, Spong CY, Brock JW, Burrows PK, Johnson MP, Howell LJ, Farrel JA, Dabrowiak ME, Sutton LN, Gupta N, Tulipan NB, D’Alton ME, Farmer DL. A randomized trial of prenatal versus postnatal repair of myelomeningocele. N Engl J Med 2011; 364(11) : 993-1004. 2. Bruner JP, Tulipan NE, Richards WO. Endoscopic coverage of fetal open myelomeningocele in utero. Am J Obstet Gynecol 1997; 176 : 256-257. 3. Bruner JP, Tulipan NB, Richards WO, Walsh WF, Boehm FH, Vrabcak EK. In utero repair of myelomeningocele: a comparison of endoscopy and hysterotomy. Fetal Diagn Ther 2000; 15 : 83-88. 4. Farmer DL, von Koch CS, Warwick JP, Danielpour M, Gupta N, Lee H, Harrison MR. In utero repair of myelomeningocele. Arch Surg 2003; 138 : 872-878. 5. Kohl T, Witteler R, Strümper D, Gogarten W, Asfour B, Reckers J, Merschhoff G, Marcus AE, Weyand M, Van Aken H, Vogt J, Scheld HH. Operative techniques and strategies for minimally invasive fetoscopic fetal cardiac interventions in sheep. Surg Endosc 2000; 14 : 424-430. 6. Kohl T, Szabo Z, Suda K, Harrison MR, Quinn TM, Petrossian E, Hanley FM. Percutaneous fetal access and uterine closure for fetoscopic surgery - Lessons from 16 consecutive procedures in pregnant sheep. Surgical Endoscopy 1997; 11 : 819-824. 7. Kohl T, Tchatcheva K, Weinbach J, Hering R, Kozlowski P, Stressig R, Gembruch U. Partial amniotic carbon dioxide insufflation (PACI) during minimally invasive fetoscopic surgery – Early clinical experience. Surg Endosc. 2010; 24 : 432-444. 8. Kohl T, Große Hartlage M, Kienitz D, Westphal M, Buller T, Aryee S, Achenbach S, Gembruch U, Brentrup A. Percutaneous fetoscopic patch coverage of experimental lumbosacral full- thickness skin lesions in sheep - A minimally invasive technique aimed at minimizing maternal trauma from fetal surgery for myelomeningocele. Surg Endoscopy 2003; 17 : 1218-1223. 9. Kohl T, Hering R, Heep A, Schaller C, Meyer B, Greive C, Bizjak G, Buller T, Van de Vondel P, Gogarten W, Bartmann P, Knöpfle G, Gembruch U. Percutaneous fetoscopic patch coverage of spina bifida aperta in the human – Early clinical experience and potential. Fetal Diagn Ther 2006; 21 : 185193. 10. Kohl T, Tchatcheva K, Merz W, Wartenberg HC, Heep A, Müller A, Franz A, Stressig R, Willinek W, Gembruch U. Percutaneous fetoscopic patch closure of spina bifida aperta in the human –
This article is protected by copyright. All rights reserved
Advances in fetal surgical techniques may now obviate the need for early postnatal neurosurgical intervention. Surg Endosc 2009; 23 : 890-895. 11. Verbeek RJ, Heep A, Maurits NM, Cremer R, Hoving EW, Brouwer OF, van der Hoeven JH, Sival DA. Fetal endoscopic myelomeningocele closure preserves segmental neurological function. Dev Med Child Neurol. 2012; 54 : 15-22. 12. Kohl T. Stool contamination. Neurosurg Pediatrics 2010; 5 : 422. 13. Kohl T, Kawecki A, Degenhardt J, Axt-Fliedner R. Preoperative sonoanatomic examination of fetal spina bifida permits prediction of surgical complexity during subsequent minimally-invasive fetoscopic closure. Ultrasound Obstet Gynecol 2012; 40 : 8. 14. Kohl T, Hering R, Van de Vondel P, Tchatcheva K, Berg C, Bartmann P, Heep A, Franz A, Müller A, Gembruch U. Analysis of the step-wise clinical introduction of experimental percutaneous fetoscopic surgical techniques for upcoming minimally-invasive fetal cardiac interventions. Surgical Endosc 2006; 20 : 1134-1143. 15. Kohl T, Ziemann M, Weinbach J, Tchatcheva K, Hasselblatt M. Partial amniotic carbon dioxide insufflation (PACI) during minimally invasive fetoscopic interventions seems safe for the fetal brain in sheep. J Laparoendosc Surg 2010; 20 : 651-653. 16. Kohl T, Reckers J, Strümper D, Grosse Hartlage M, Gogarten W, Gembruch U, Vogt J, Van Aken H, Scheld HH, Paulus W, Rickert CH. Amniotic air insufflation during minimally invasive fetoscopic fetal cardiac interventions is safe for the fetal brain in sheep. J Thorac Cardiovasc Surg. 2004 ; 128 : 467-471. 17. Luks FI, Deprest JA, Marcus M, Vandenberghe K, Vertommen JD, Lerut T. Carbon dioxide pneumamniosis causes acidosis in fetal lambs. Fetal Diagn Ther 1994; 9 : 105-109. 18. Gratacos E, Wu J, Devlieger R, Van de Velde M, Deprest JA. Effects of amniodistension with carbon dioxide on fetal acid-base status during fetoscopic surgery in the sheep model. Surg Endosc 2001; 15 : 368-372. 19. Sayki Y, Litwin DE, Bigras JL, Waddel J, Konig A, Baik S, Navsarikar A, Rebeyka IM. Reducing the deleterious effects of intrauterine CO2 during fetoscopic surgery. J Surg Res 1997; 69 : 51-54. 20. Hering R, Hoeft A, Putensen C, Tchatcheva K, Stressig R, Gembruch U, Kohl T. Maternal haemodynamics and lung water content during percutaneous fetoscopic interventions under general anaesthesia. Brit J Anaesth 2009; 102 : 523-527.
This article is protected by copyright. All rights reserved
21. Meuli M, Moehrlen U. Fetal surgery for myelomeningocele: a critical appraisal. Eur J Pediatr Surg. 2013; 23 : 103-109. 22. Herrera SR, Leme RJ, Valente PR, Caldini EG, Saldiva PH, Pedreira DA. Comparison between two surgical techniques for prenatal correction of meningomyelocele in sheep. Einstein (Sao Paulo). 2012; 10 : 455-461. 23. Aaronson OS, Tulipan, NB, Cywes R, Sundell HW, Davis GH, Bruner JP, Richards WO. Robotassisted endoscopic intrauterine myelomeningocele repair: a feasibility study. Pediatr Neurosurg 2002; 36 : 85-89. 24. Shurtleff D. Fetal endoscopic myelomeningocele repair. Dev Med Child Neurol. 2012; 54 : 4-5. 25. Kohl T. Iatrogenic fetal membrane damage from complex fetoscopic surgery in human fetuses might not be amenable to simple closure by collagen plugs. Prenat Diagn 2008; 28 : 876-877. 26. Engels AC, Van Calster B, Richter J, Dekoninck P, Lewi L, De Catte L, Devlieger R, Deprest JA. Collagen plug sealing of iatrogenic fetal membrane defects after fetoscopic surgery for congenital diaphragmatic hernia. Ultrasound Obstet Gynecol 2013. DOI: 10.1002/uog.12547.
This article is protected by copyright. All rights reserved
ID Age Lesion level
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
31 30 23 31 36 36 22 34 29 38 35 30 19 36 30 30 30 19 41 40 38 36 29 34 40 25 28 28 33 36
Placental position Gestational age at fetal surgery Gestational age at delivery ( Anterior / Posterior) skin-to-skin-time weeks+days weeks+days P 24+0 34+2 230 A 22+6 29+4 240 A 25+0 34+0 240 A 24+0 36+5 255 A 23+4 24+4 270 P 21+1 34+0 240 A 25+0 30+4 290 P 24+6 30+0 300 A 25+3 30+4 260 A 21+5 32+3 195 A 25+3 30+5 265 P 27+2 36+4 270 P 22+0 34+3 180 P 23+3 34+1 210 P 25+1 32+0 315 P 24+4 38+1 255 P 22+1 34+3 270 P 24+1 31+0 240 P 22+6 30+0 300 A 22+4 34+3 210 A 24+6 33+4 210 A 21+4 28+5 215 P 23+0 30+0 210 A 21+3 33+0 190 P 23+3 37+2 210 P 25+4 34+1 225 P 22+2 34+6 225 P 22+4 37+2 225 P 24+4 31+3 245 P 23+1 29+6 245
This article is protected by copyright. All rights reserved
L4 L1 L5 L4 L3 L2 L2 L5 L5 L4 L3 S1 L4 L4 L4 L5 L1 L4 L1 L4 L5 L3 L4 L3 L5 L3 L4 L4 L3 L5
Lesion type flat / cystic f f f f f f f f c f f f c c c f f f f f f f f f f c f c f f
Lesionsize (cm) Amniotic access achiev 3 4x3 4x3 4x3 4x3 4x3 4x2 3 2 2 x 1.5 3.5 x 3 3x2 3 3 3 2 x 1.5 4x3 3 x 1.5 3 x 3.5 2.5 x 3 3x2 2 x 1.5 2 2.5 2.5 4.5 x 3 4 x 2.5 4 3 x 2.5 3 x 2.5
+ + + + + + + + + + + + + + + + + + + + + + + + + + + + + +
31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51
34 23 30 40 33 29 26 23 37 25 26 32 32 30 38 37 32 28 32 30 28
P A A A P P P P A A P A P P P P P P A P A
24+0 24+4 23+6 21+0 26+6 21+4 22+4 22+6 24+3 23+1 28+5 22+5 24+0 29+1 22+3 23+0 23+2 23+6 22+4 23+1 23+4
29+2 32+5 35+1 31+1 38+1 34+2 33+6 32+2 30+2 34+2 34+0 32+1 30+6 36+0 34+2 34+4 31+0 34+1 27+6 35+1 34+1
205 45 240 165 160 150 155 200 225 180 140 220 200 210 185 190 190 225 280 175 225
L1 L3 L5 L4 L5 L5 L5 T 11 L5 L2 L5 L5 Th 11 L4 L4 L3 L3 S1 L2 L5 L2
f f f c f c f f c f f f f f c c c f c c c
3 x 2.5 n.a. 3 x 2.5 3 x 2.5 3x2 2 3 x 2.5 4x3 4x3 5x3 4 x 3.5 3.5 cm 5x4 4x3 5x4 5x4 3.5 x 3 3.5 x 3 4 x 3.5 3x2 3.5 x 3
- n.a. = not applicable; DC = direct closure; DCP = direct closure over patch; DP = double patch closure; SP = single patch closure; TP = triple patch closure
This article is protected by copyright. All rights reserved
+ + + + + + + + + + + + + + + + + + + + +
Legends
Figure 1: External aspect of the percutaneous setup - Percutaneous minimally-invasive fetoscopic surgery for SBA is performed via three (to four) trocars with an external diameter of 5 mm that are placed into the amniotic cavity under continuous ultrasound monitoring by a Seldinger approach (top). Following partial amniotic carbon dioxide insufflation (PACI) and fetal posturing, surgical dissection of the malformation is carried out using fetoscopic instruments (bottom).
This article is protected by copyright. All rights reserved
Figure 2: Minimally-invasive fetoscopic double patch closure of a large L3-myelomeningocele at 23+2 weeks of gestation. Top row – left – The large myelomeningocele before surgical dissection. The exposed neural tissue can be visualized clearly in the gas-insufflated amniotic cavity. Top row right - The malformation is being circumcised with a needle electrode. Middle row – left - Note the high degree of precision that can be achieved within less than a millimeter range during manual dissection of the placode using micro-grasper and micro-scissors. Middle row – right: Fetoscopic view of the neural tissue after removal of surrounding pathological tissues. Note the excellent visualization of the spinal cord nerves in the high-powered fetoscopic field employing a 3.3 mm-30° rod-lens fetoscope. Bottom row – left: Fetoscopic view after protecting the neural tissue with an inert Gore⎝-PTFE patch. This step aims at the prevention of tethering. Bottom row – right: Definitive coverage of the entire surgical field is then achieved with a collagen patch. The goal of watertight coverage is demonstrated at the end of the procedure by bulging of the patch as well as lack of cerebrospinal fluid leakage when compressed with an instrument.
This article is protected by copyright. All rights reserved
Figure 3: Minimally-invasive fetoscopic direct closure of a sacral myeloschisis at 23+6 weeks of gestation. Top row – left – The sacral myeloschisis before surgical dissection. A small amount of stool (arrow) can be visualized over the opening of the central canal at the rostral end of the neural tissue. Top row - right – Following surgical dissection, the malformation has been covered with a collagen patch. Bottom row – left – Then, complete skin closure has been achieved. Bottom row – right: After delivery, only a very small part of the collagen patch can be seen as a result of dehissence of one of the skin sutures. The skin overgrew the patch without the need for any further intervention. Figure 4: Top - Employing ultrasound-guidance, percutaneous fetal access with three trocars can be achieved in essentially all cases. As each trocar is inserted over a wire guide, the intial puncture hole for maternal percutaneous-transabdominal-transuterine-intraamniotic trocar insertion has a diameter of only 1.2 mm. Bottom – The tiny incisions and punctures for trocar insertion during the minimally-invasive approach are barely comparable to the much larger incisions in the maternal abdomen and uterus required for open fetal surgery.
This article is protected by copyright. All rights reserved
This article is protected by copyright. All rights reserved
Figure 5: Top – Newborn with spina bifida at 35 weeks of gestation following double-patch closure of an L4 defect extending over the entire sacrum at 22+4 weeks of gestation. As in human fetuses the surface of the patch does not overgrow with skin over the remainder of gestation, this step has to happen within the first postnatal weeks of life. Middle – The perfectly ingrown upper patch after removal of the surgical clips. Bottom – Fetoscopic image of the large malformation with a long segment of exposed placode (arrow) before closure. Note that at the time of surgery, many malformation are almost as wide as the entire fetal back.
This article is protected by copyright. All rights reserved
This article is protected by copyright. All rights reserved
Figure 6: Surgically challenging setups for minimally-invasive fetoscopic surgery of spina bifida that may benefit from the incorporation of surgical robotics. Top left – Large L4-myelomeningocele with a terminal myelocystocele (m) at 25+1 weeks of gestation. The main surgical difficulty resulted from sparing the spinal nerves (arrows) that were attached along the inner surface of the sac from injury during resection of the sac tissue. Top right – after removal of most parts of the sac, the intact nerves (arrows) can be seen still adherent to a part of the inner layer of the sac that was preserved. Middle left - Large L5-myelomeningocele. The delicate nerves of the cauda equina (ce) are found attached (arrow) to the far sturdier pathological skin at 24+3 weeks of gestation. Middle right – Fetoscopic image after removal of all sac tissue. Bottom left – L5-myelomeningocele with pathologic skin adjacent to the neural placode (p) at 26+6 weeks of gestation. Beige-colored stool can be seen along the placode and also on its surface. As in the case depicted in the middle row, the surgical difficulty here results from the dissection of the delicate neural tissue from the far sturdier pathological skin. Bottom right – In this case, a firm strand of amniotic tissue divides the amniotic cavity in two compartments. This problem makes posturing maneuvers more difficult and in itself poses a risk factor for preterm rupture of membranes.
This article is protected by copyright. All rights reserved
This article is protected by copyright. All rights reserved
