Surg Endosc DOI 10.1007/s00464-013-2984-6
and Other Interventional Techniques
Single-Access Fetal Endoscopy (SAFE) for myelomeningocele in sheep model I: amniotic carbon dioxide gas approach Jose L. Peiro • Cesar G. Fontecha • Rodrigo Ruano Marielle Esteves • Carla Fonseca • Mario Marotta • Sina Haeri • Michael A. Belfort
•
Received: 16 February 2013 / Accepted: 5 April 2013 Ó Springer Science+Business Media New York 2013
Abstract Background This study aimed to assess the feasibility of single-access fetal endoscopy (SAFE) for the management of myelomeningocele (MMC) using intrauterine carbon dioxide as a distension medium in a sheep model. Methods This prospective experimental case-control study investigated 12 lamb fetuses that had a myelomeningocele-like defect surgically created on the 75th day of gestation. Four fetuses remained untreated (control group), and eight fetuses had MMC repair using two fetoscopic approaches with carbon dioxide used to distend the amniotic cavity. A collagen patch was placed over the defect and secured with surgical sealant. Four animals had a twoport fetoscopic procedure, and four animals had SAFE. Clinical and pathologic studies were performed after delivery. Results This study confirmed the validity of the animal MMC model. None of the control animals was able to stand or walk, and all had a significant defect in the lumbar area with continuous leakage of cerebrospinal fluid, ventriculomegaly, and a Chiari-II malformation. All the treated animals, independently of the number of ports used in the repair, were able to walk and had a closed defect with resolution of the Chiari malformation.
J. L. Peiro (&) C. G. Fontecha M. Esteves C. Fonseca M. Marotta Fetal Surgery Program, Congenital Malformations Research Group, Research Institute of Hospital Universitari Vall d’Hebron, Edifici Infantil, Passeig Vall d’Hebron 119-129, 08035 Barcelona, Spain e-mail:
[email protected] R. Ruano S. Haeri M. A. Belfort Department of Obstetrics and Gynecology, Texas Children’s Fetal Center, Baylor College of Medicine, Houston, TX, USA
Conclusions The SAFE patch and glue coverage of surgically created fetal MMC is feasible and effective in restoring gross neurologic function in the fetal lamb model. Keywords Animal model Fetal surgery Fetoscopy Myelomeningocele Single access Spina bifida
Fetal myelomeningocele (MMC) is a devastating neurologic congenital anomaly characterized by a closure defect of the spinal column [1–3]. A recent randomized trial demonstrated that open fetal surgery is effective in reducing the postnatal neurologic morbidity, as evidenced by decreased incidence and severity of postnatal hydrocephalus and reduced need for postnatal ventricular-peritoneal shunting [4]. However, open fetal surgery is associated with increased potential for maternal–fetal morbidity including prematurity, rupture of the membranes, uterine rupture and dehiscence, maternal hemorrhage, and hysterectomy. A fetoscopic approach potentially offers a less invasive therapeutic option, with the aim of reducing both maternal and fetal morbidity. Both animal studies and isolated case reports of clinical human experience with a fetoscopic approach to managing MMC have been reported in the literature. These have demonstrated the feasibility of covering an MMC defect with patch and suture or sealant but have involved using at least two or more access ports into the uterus [5–11]. We have previously described an approach using simple intrauterine fetal MMC coverage [5], as opposed to definitive surgical repair using a patch and glue technique, which is specifically amenable to a fetoscopic approach [6]. Given that the ideal is to perform intrauterine fetoscopic procedures using a technique that minimizes maternal and fetal risks, we aimed in the current study to evaluate the
123
Surg Endosc
feasibility of single-access-port fetal endoscopic repair (SAFE) in a sheep model of MMC.
Materials and methods This experimental study was performed according to European Council Directives (C86/609/EEC and 2003/65/ EC) and Spanish guidelines for the use of laboratory animals (Boletı´n Oficial del Estado of 18 March 1988) and with the approval of the local Ethics Committee for experimental animal use. Healthy young pregnant ewes (Ripollese breed) were transferred to our animal facilities 10 days before the planned intervention and allowed free access to food and water except during the 24 h before surgery. An ultrasound study was performed before the procedures to confirm the viability of the pregnancy and to document the gestational age. A surgical MMC defect was created on day 75 of gestation using the method described in our previously published protocol [5, 6]. Briefly, an MMC was created on gestational day 75 (full term, 145–150 days). Pregnant ewes were premedicated using 0.5 mg/kg intravenous (IV) midazolam plus 0.02 mg/kg IV buprenorphine and anesthetized using 5 mg/kg IV propofol plus 3 % inhaled isoflurane. The animals underwent surgery in the supine position. An infraumbilical midline laparotomy was performed to expose the uterus. The uterus was opened, and the fetal spine was brought to the opening. A lumbar defect was created by resecting the skin and paraspinal muscles, performing a laminectomy of three lumbar vertebrae, and opening the dura mater. After the MMC had been surgically created, the fetal lambs were randomly divided into the following three groups: untreated control group, two-port fetoscopic coverage group (TPFE), and SAFE. The animals in the TPFE and SAFE groups underwent a second procedure at the gestational age of 95 days (Fig. 1). Both fetoscopic techniques were performed after exposure of the utero through longitudinal midline laparotomy. The two-port fetoscopic coverage of the MMC defect was performed according to the technique previously described [6]. Briefly, the fetal spine was gently positioned and held in an anteroposterior orientation without opening of the uterus. Two vascular catheters then were inserted into the amniotic cavity: (1) a 12-Fr Cook catheter (Cook Group Inc., Bloomington, IN, USA) at the level of the fetal cervical spine, angulated toward the lumbar area, through which the 2-mm fetoscope was inserted, and (2) a 10-Fr Cook catheter (Cook Group Inc., Bloomington, IN, USA) for the insertion of instruments used to deliver and place the patch and sealant.
123
Using the 12-Fr trocar, a small amount of amniotic fluid was removed and set aside for injection back into the uterus at completion of the procedure. Carbon dioxide then was insufflated into the uterus through the 12-Fr access port and maintained at a maximum pressure of 5 mmHg. An appropriately sized Silastic patch (Sumsa, Madrid, Spain) or a collagen patch (MatridermÒ; Dr. Suwelack Skin & Health Care AG, Billerbeck, Germany) was inserted into the uterus through the 10-Fr access port and manipulated into position, using grasping forceps (26159UHW, Karl Storz, Germany), over the spinal defect. A surgical sealant (Coseal Surgical Sealant; Baxter Healthcare Corp., Deerfield, IL, USA) then was dripped onto the patch such that the entire patch and adjacent skin were covered. The sealant was allowed to cure for 3 min, after which the gas was removed and the amniotic fluid returned to the amniotic cavity. The two small access ports and the abdominal wall incision then were closed. The SAFE procedure was performed via a single-access port using a 12-Fr vascular catheter (Cook Group Inc., Bloomington, IN, USA) introduced into the amniotic cavity using the Seldinger technique under ultrasound guidance. The uterus was exposed via a maternal laparotomy, and the fetal back was positioned in the same way as for the two-port technique. The 12-Fr cannula was introduced directly over the posterior fetal neck and then angulated toward the spina bifida defect (Fig. 2). Using the same technique, a small amount of amniotic fluid was removed (*1/3 of the total) and replaced with carbon dioxide at the lowest pressure that allowed adequate visualization of the defect (*5 mmHg). A 2.7-mm fetoscope with a 1-mm working channel was advanced through the cannula (Fig. 3A), allowing visualization of the fetal back and the spinal defect, which was above the level of the amniotic fluid. The lamb fetus was floating back-up in the amniotic fluid (Fig. 3B). After the spina bifida defect had been identified, the fetoscope was removed from the cannula, and a small collagen-elastine patch (MatridermÒ; Dr. Suwelack Skin & Health Care AG, Billerbeck, Germany), previously cut into an appropriately sized ovoid shape and rolled up to fit into the cannula, was inserted into the cannula and pushed down toward the lower half of the cannula. The fetoscope with a 1-mm gasping forceps (26159UHW, Karl Storz, Germany) protruding from the working channel then was introduced into the upper portion of the cannula such that the rolled-up patch could be seen and grasped. Under vision, the patch then was advanced into the amniotic cavity and gently placed on the fetal back. Using the grasping forceps, the patch was unrolled and then moved into position over the spinal defect (Fig. 3C, D) such that the open neural tube was completely covered. The 1-mm forceps then was removed, and a 1-mm catheter was inserted into the
Surg Endosc Fig. 1 Flow diagram of the study
Fig. 2 Scheme of the single-access fetal endoscopy (SAFE) procedure in the sheep model
working channel and used to deliver the Sealant (CosealÒ Surgical Sealant; Baxter Healthcare Corp.) onto the patch and the adjacent skin (Fig. 3E, F). After 3 min (to allow glue polymerization), the carbon dioxide was aspirated, and the previously removed and stored amniotic fluid was replaced. The 12-Fr vascular access catheter was removed; the small uterine hole was sutured closed; and the maternal abdominal wall was closed. The lamb fetuses were delivered by cesarean section on days 135–140 of gestation and placed under veterinary monitoring. At 48 h of life, the animals were clinically evaluated by assessment of the following parameters: hind-leg paraplegia (ability to stand unaided), sphincter continence (urine loss when pressure was applied over the bladder), foreleg ataxia (abnormally wide support base during standing), and
swallowing difficulty during feeding. In addition, the region of the spina bifida defect was observed for macroscopic cerebrospinal fluid (CSF) loss and percentage of closure. After the lamb had been killed, one carotid artery was cannulated, and the animal was perfused with 10 % formaldehyde. Both the spinal cord structures and the brain were macro- and microscopically studied. The two surgical approaches (single port and double port) were evaluated with respect to the following parameters: fetal loss after each surgery, duration of the procedure, degree of difficulty of the fetal intervention [using a standardized scale of 0 (very easy) to 5 (very difficult)], maximum carbon dioxide pressure, success regarding defect coverage, and perioperative fetal–maternal complications. Based on a power of 0.8 to detect a 90 % difference in the closure of the defect between the treated and untreated animals, at least four animals were required per group. Statistical analyses were performed using the chi-square test or Fisher’s exact test, as appropriate, and the Mann-Whitney U test. Statistical significance was set at a p \ 0.05.
Results A significant MMC defect was created in 14 lamb fetuses on day 75 of gestation. In all cases, leakage of CSF was confirmed, and the spinal cord was adequately exposed. Two fetuses died after the creation of the MMC defect. Fetoscopic coverage of the defect was performed in eight fetuses on day 95 of gestation. Four animals underwent TPFE, and four lamb fetuses underwent SAFE (Fig. 3). The two groups did not differ statistically with respect to the duration of the procedure, the carbon dioxide pressure, or any of the other measured parameters, as shown in Table 1.
123
Surg Endosc Fig. 3 Single-access fetal endoscopy (SAFE) myelomeningocele coverage procedure. A Only one port is placed in the ovine exposed uterus. B Fetoscopic view of the spinal defect in a sheep model. C Covering of the spinal defect with the patch using an intrauterine 1-mm forceps. D Spinal defect completely covered by the patch. E Delivery of the sealant over the patch. F Final aspect of the spinal defect covered by the patch and sealant
All the fetuses were delivered by planned cesarean section on days 135–140 of gestation. All the untreated animals had an open spinal defect with CSF leakage, a Chiari malformation, and paraplegia. None of the surgically repaired animals had CSF leakage, Chiari
malformation, or paraplegia, and all had almost complete epithelialization and closure of the defect at birth. Histologic studies confirmed severe damage of the spinal cord tissue in the untreated group, whereas in contrast, both groups of surgically repaired animals showed a
Table 1 Characteristics of lamb fetuses that underwent two-port or single-access fetoscopic coverage of a myelomeningocele defect
Procedure duration (min) Carbon dioxide pressure (mmHg)
Two-port fetoscopy (n = 4)
Single-port fetoscopy (n = 4)
p value
26.2 ± 7.2 4.3 ± 0.8
28.4 ± 8.4 4.5 ± 0.7
0.20 0.99
Difficulty of surgical procedure (1–5): n (range)
2 (2–3)
3 (2–4)
0.21
Surgical coverage success: n (%)
4 (100)
4 (100)
1.00
Fetal loss: n (%)
0
0
1.00
Surgical complications: n (%)
0
0
1.00
123
Surg Endosc
normal-appearing spinal cord covered with newly formed dura mater as well as disordered muscle fibers and subcutaneous-like tissue and skin.
Discussion The current study demonstrated the surgical feasibility of SAFE repair in a carbon dioxide environment of a surgically created spina bifida defect using a simple patch and glue technique. We have previously demonstrated successful two-port fetoscopic repair of an MMC-like defect in a sheep model using a similar patch and glue technique [6]. The current study has taken this approach one step further by showing that an equally successful closure of the defect can be achieved with a single-port approach. We believe that by reducing the number of ports needed and by simplifying the surgical technique, the duration and complexity of the surgery can be reduced, with a concomitant decrease in maternal complications. In 1999, Bruner et al. [10, 11], first introduced fetoscopic repair of MMC using a three-port technique. This approach was subsequently abandoned because of technical difficulties and lack of benefit. A number of experimental studies in the sheep model with patch and suture techniques have been published [5, 7, 11]. Human data on fetoscopic repair of MMC using a sutured patch technique also have been published [8, 12]. All these studies, however, have been accomplished using three or more access ports into the uterus, with prolonged surgical time related to the suturing process [8, 12]. We have previously demonstrated reversal of CSF leakage and Chiari malformation with successful and effective coverage of a surgically created defect using a patch and sealant technique that led to improved neurologic outcomes [6, 13]. To our knowledge, the current study is the first to report on the feasibility of performing fetoscopic coverage of a spinal defect using a patch and sealant technique in an amniotic cavity carbon dioxide environment using single-port access. This approach had results similar to those obtained with a two-port-access technique, with no statistical differences in the duration of the procedure, the required carbon dioxide pressure, or the perceived difficulty of the surgery. In addition, the two-port- and single-port-access groups did not differ in terms of postoperative skin coverage or prevention of subsequent neurologic problems secondary to the spinal defect [6]. The advantage of the single-accessport technique relates to the potential reduction in the risk for preterm rupture of the membranes and prematurity, a risk that may be associated with the number of ports and the duration of the procedure. Given our results with the animal model, we speculate that the SAFE repair technique in a low-pressure amniotic
cavity carbon dioxide environment may be a therapeutic option for human fetuses with MMC. Kohl et al. [8, 14] have reported their experience with fetoscopic coverage of MMC in humans using a patch and suture technique. We believe that given the animal data, a study of the SAFE patch and glue repair technique in humans is now ethically justified. In conclusion, single-access fetoscopic coverage of MMC under an amniotic carbon dioxide medium is feasible in a sheep model and may be a therapeutic alternative for human fetuses with MMC that offers the potential to reduce maternal and fetal risks. Disclosures Jose L. Peiro, Cesar G. Fontecha, Rodrigo Ruano, Marielle Esteves, Carla Fonseca, Mario Marotta, Sina Haeri, and Michael A. Belfort have no conflicts of interest or financial ties to disclose.
References 1. Au KS, Ashley-Koch A, Northrup H (2010) Epidemiologic and genetic aspects of spina bifida and other neural tube defects. Dev Disabil Res Rev 16:6–15 2. Amari F, Junkers W, Hartge D, Beyer DA, Axt-Fliedner R, Weichert J (2010) Prenatal course and outcome in 103 cases of fetal spina bifida: a single-center experience. Acta Obstet Gynecol Scand 89:1276–1283 3. Adzick NS (2010) Fetal myelomeningocele: natural history, pathophysiology, and in utero intervention. Semin Fetal Neonatal Med 15:9–14 4. Adzick NS, Thom EA, Spong CY, Brock JW III, Burrows PK, Johnson MP, Howell LJ, Farrell JA, Dabrowiak ME, Sutton LN, Gupta N, Tulipan NB, D’Alton ME, Farmer DL, MOMS Investigators (2011) A randomized trial of prenatal versus postnatal repair of myelomeningocele. N Engl J Med 364:993–1004 5. Fontecha CG, Peiro JL, Aguirre M, Soldado F, An˜or S, Fresno L, Martinez-Iban˜ez V (2009) Inert patch with bioadhesive for gentle fetal surgery of myelomeningocele in a sheep model. Eur J Obstet Gynecol Reprod Biol 146:174–179 6. Fontecha CG, Peiro JL, Sevilla JJ, Aguirre M, Soldado F, Fresno L, Fonseca C, Chacaltana A, Martinez V (2011) Fetoscopic coverage of experimental myelomeningocele in sheep using a patch with surgical sealant. Eur J Obstet Gynecol Reprod Biol 156:171–176 7. Pedreira DA, Sanchez-Oliveira RC, Valente PR, Abou-Jamra RC, Araujo A, Saldiva PH (2007) Validation of the ovine fetus as an experimental model for the human myelomeningocele defect. Acta Cir Bras 22:168–173 8. Kohl T, Tchatcheva K, Merz W, Wartenberg HC, Heep A, Mu¨ller A, Franz A, Stressig R, Willinek W, Gembruch U (2009) Percutaneous fetoscopic patch closure of human spina bifida aperta: advances in fetal surgical techniques may obviate the need for early postnatal neurosurgical intervention. Surg Endosc 23:890–895 9. Copeland ML, Bruner JP, Richards WO, Sundell HW, Tulipan NB (1993) A model for in utero endoscopic treatment of myelomeningocele. Neurosurgery 33:542–544 10. Bruner JP, Richards WO, Tulipan NB, Arney TL (1999) Endoscopic coverage of fetal myelomeningocele in utero. Am J Obstet Gynecol 180(1 Pt 1):153–158 11. Bruner JP, Tulipan NB, Richards WO, Walsh WF, Boehm FH, Vrabcak EK (2000) In utero repair of myelomeningocele: a
123
Surg Endosc comparison of endoscopy and hysterotomy. Fetal Diagn Ther 15:83–88 12. Encinas JL, Garcı´a-Cabezas MA, Barkovich J, Fontecha CG, Peiro´ JL, Soto GM, Borrell V, Reillo I, Lo´pez-Santamarı´a M, Tovar JA, Farmer DL (2011) Maldevelopment of the cerebral cortex in the surgically induced model of myelomeningocele:
123
implications for fetal neurosurgery. J Pediatr Surg 46: 713–722 13. Peiro´ JL, Encinas JL (2012) Fetal surgery of myelomeningocele: from experimental results to clinical evidence. Cir Pediatr 25:1–3 14. Kohl T (2010) Minimally invasive fetoscopic interventions: an overview in 2010. Surg Endosc 24:2056–2067
