Fetal endoscopic tracheal occlusion (‘Fetendo-PLUG’) for congenital diaphragmatic hernia

Journal VOL31,NO

of Pediatric

Surgery OCTOBER 1996

10

Fetal Endoscopic Tracheal Occlusion (‘Fetendo-PLUG’) for Congenital Diaphragmatic Hernia By Erik D. Skarsgard,

Martin

Meuli,

Karen J. VanderWall, John F. Bealer, San Francisco. California

0 Despite recent advances in surgical technique, posthysterotomy preterm labor remains a major determinant of postoperative fetal morbidity and mortality after in utero repair of congenital diaphragmatic hernia (CDH). Temporary fetal tracheal occlusion, or “PLUG” (Plug the Lung Until it Grows), reverses the pulmonary hypoplasia seen in experimental models of CDH and provides an alternative treatment strategy for some fetuses with CDH. Adaptation of current, minimally invasive surgical technology to the PLUG technique allows treatment of CDH without opening the uterus. In this report the authors describe a video-fetoscopic, intrauterine technique of tracheal occlusion (called FetendoPLUG) that could be used in human fetuses with CDH. The technique was developed in four fetal lambs that underwent video-fetoscopic intervention at 110 days’ gestation1 (full term, 145 days), having undergone open creation of diaphragmatic hernias at 75 days. After maternal laparotomy and uterine exposure, the fetal head was located and a 5-mm curved, balloon-cuffed trocar was introduced through a uterine puncture directly into the fetal oral cavity. A steerable “bronchoscope” (with an instrument channel) was used to endoscopically intubate the trachea through the trocar, and the trocar was advanced over the bronchoscope and its balloon inflated to provide secure tracheal access below the vocal cords. Next, a IO-mm trocar was placed directly over the fetal neck, and the amniotic space was expanded with warm saline. A 5-mm laparoscope was introduced, and under simultaneous, dual video-fetoscopic (endotracheal and endoamniotic) visualization, a l-mm nephrostomy puncture wire was advanced along the instrument channel of the bronchoscope, through the anterior wall of the trachea and fetal neck, into the amniotic space, then through the uterine wall to the outside. Withdrawal of the bronchoscope over the wire left a 5-mm endotracheal “trocar channel” along which a compressed, gelatin-encapsulated, polymeric foam insert (outer diameter, 4.8 mm) could be delivered by suture attachment to the guide wire. Once the foam was in its final endotracheal position, dissolution of the gelatin membrane allowed expansion of the foam to produce a water impervious tracheal occlusion This two-trocar video-fetoscopic PLUG technique was performed successfully in all four fetuses, with a sequential decrease in operating time (median, 3.5 hours). Although two fetuses aborted postoperatively, the other two were carried successfully to term and demonstrated the anticipated physiological effects of adequate tracheal occlusion at the time of delivery. Copyright o 1996 by W. B. Saunders Company INDEX WORDS: fetoscopy, tracheal JournalofPediatricSorgery,

Congenital occlusion,

diaphragmatic fetal surgery.

's'ol31,No

hernia,

10 (October),1996:

video-

pp 1335.1338

N. Scott Adzick,

and Michael

R. Harrison

T

HE HIGH postnatal mortality rate from pulmonary hypoplasia associated with congenital diaphragmatic hernia (CDH) diagnosed before 24 weeks’ gestation has led to the evolution of in utero techniques that promote prenatal pulmonary growth.’ Although in utero diaphragmatic hernia repair permits fetal pulmonary growth, the hysterotomy necessary for fetal exposure causes increased fetal morbidity and mortality through postoperative preterm labor.2 Because the risk of preterm labor appears to be closely related to the size of the hysterotomy, a “minimally invasive” treatment strategy should increase significantly the safety of prenatal surgical therapy for fetuses with CDH. Recently we reported our experience with in utero tracheal occlusion, or “PLUG” (Plug the Lung Until it Grows), in a 27-week human fetus with CDH.3 By adapting minimally invasive video-fetoscopy to our current technique of fetal tracheal occlusion, we have developed a means of treating experimental CDH without opening the uterus. This new experimental video-fetoscopic technique (which we call FetendoPLUG) is the focus of this report. MATERIALS

AND METHODS

Creation of Diaphragmatic Hernia Four time-dated gravid ewes (Torrel Farms, Ukiah, CA) underwent fetal surgery at 75 days’ gestation (full term, 145 days) for creation of fetal diaphragmatic hernias, as described previously.4

From the Fetal Treatment Center, University of California, San Francisco, CA. Supported in part by the Hospital For Sick Children Foundation, Toronto, Ontario, Canada. Address reprint requests to Michael R. Hamson, MD, Fetal Treatment Center, University of California, San Francisco, 1601 HSq 513 Pamassus Ave, San Francisco, CA 94143-0570. Copyright o 1996 by WB. Saunders Company 0022-3468196131IO-OOOl$O3.OOlO

1335

SKARSGARD

Guidewe

In brief, under halothane anesthesia, a maternal laparotomy and hysterotomy were performed, and the left fetal hemithorax was entered through the ninth interspace. The exposed left hemidiaphragm was disrupted by a cruciate incision, and the stomach and small bowel were delivered manually into the chest. After fetal chest closure and restoration of amniotic fluid volume, the hysterotomy was repaired with a TA-90 stapler, (US Surgical Co, Norwalk, CT), and the maternal abdomen was closed.

Edeo-Fetoscopic (Fetendo-PLUG)

Tracheal Occlusion

This intervention was performed at approximately 110 days’ gestation. Under general anesthesia with inhaled halothane, the gravid ewes underwent a second laparotomy for uterine exposure. The fetal head was located by palpation and externally immobilized with the mouth open. A specially adapted 5-mm curved, balloon-cuffed trocar (Innerdyne Medical Inc, Sunnyvale, CA) was introduced through a purse-string, controlled uterine puncture directly into the fetal oral cavity. A 2.8-mm steerable laparoscopic choledochoscope (EndoMedix Corp, Irvine, CA) with a 1.2-mm instrument channel was used as a bronchoscope, introduced through the trocar, and under endoscopic guidance, advanced into the trachea. The curved trocar was then advanced over the “bronchoscope,” and its cuff was inflated below the vocal cords to provide direct, secure, 5-mm access to the fetal trachea. Next, a lo-mm trocar with an inflatable, membrane-controlling balloon (Origin Medsystems Inc, Menlo Park, CA) was placed through a uterine purse string into the amniotic cavity, directly over the extended fetal neck. A 5-mm 0” endoamniotic telescope (Karl Storz Endoscopy-America, Inc, Culver City, CA) was placed through the trocar, and an amniotic fluid exchange with warmed sterile saline was performed through the trocar’s insufflation channel until subject visualization through amniotic fluid was optimal. Using a video-coupling device, simultaneous endotracheal and endoamniotic images were transmitted to the television monitor.

ET AL

Fig 1. Fetendo-PLUG technique of video-fetoscopic tracheal occlusion. (A) With secure subglottic endotracheal trocar access, the sharp guide wire is advanced through the working channel of the bronchoscope, across the anterior trachea and cervical soft tissue, and into the uterine cavity under direct endoamniotic visualization. (6) Once the bronchoscope has been withdrawn over the transtracheal guide wire, the compressed foam plug (attached via suture to the guidewire) is drawn into the trachea through the trocar. (C) With the foam plug positioned inside the trachea, the endotracheal trocar is removed and replaced with a short endoamniotic trocar through which endoscopic clips are apphed to the suture at its exit from the lamb’s neck. The suture is divided with scissors.

Under dual video-fetoscopic (endotracheal and endoamniotic) guidance, a l-mm retrograde nephrostomy puncture wire (Cook Urological, Bloomington, IN) was advanced along the instrument channel of the bronchoscope and out through the anterior tracheal wall (Fig 1A and 2). Once the puncture wire had traversed the trachea and soft tissues of the fetal neck, its exit through cervical skin and its subsequent safe course through the amniotic space to the uterine wall were easily visualized through the endoamniotic telescope. The puncture wire was advanced along this course until it had exited the uterus; then the bronchoscope was withdrawn over the wire, leaving a 5-mm endotracheal trocar channel with guide-wire access. Tracheal occlusion was achieved using a water-impermeable, polymeric foam insert with a nonabsorbable suture attached to both ends, as described previous1y.j The foam was compressed and

Fig 2. Simultaneous endotracheal and endoamniotic (inset) view of the guide wire (GW), traversing the trachea (Tr). and the cervical soft tissues and entering the amniotic cavity (AC).

FETENDO-PLUG

1337

FOR CDH

placed inside a gelatin capsule (outer diameter, 4.8 mm), and then attached (via suture) to the endotracheal guide wire (Fig 1B). Traction on the guide wire at its exit from the uterus resulted in delivery of the compressed foam through the endotracheal trocar to its final position below the vocal cords. Endoamniotic visualization (at the cervical-skin exit site) of an ink mark placed on the traction suture 1 in from the end of the plug was used to correctly position the plug within the trachea. The second suture (attached to the cephalad end of the foam) was left to exit the lamb’s mouth. At this point, the endotracheal trocar was removed and replaced with a second lo-mm endoamniotic trocar with a balloon cuff for membrane control. Through this second trocar, a medium clip applier (Ethicon, Inc, Somerville, NJ) was used to secure the traction suture at its exit from the cervical skin, thereby preventing plug migration. The guide wire was released by dividing the traction suture (distal to the site of clip fastening) with endoscopic scissors (Ethicon, Inc, Somerville, NJ) (Fig 1C). After ensuring adequate restoration of amniotic fluid volume with normal saline, and following the addition of 1 g of penicillin, the trocars were removed and the uterine trocar sites closed with 2-O silk suture. The abdomen was closed, and the ewe was awakened and returned to the stall, where food and water were provided ad libitum.

Cesarean Delivery and Resuscitation of Fetendo-PLUG Lambs Postoperative ultrasonography was performed weekly to assess fetal viability and to evaluate lung growth in response to tracheal occlusion. At 135 to 140 days, the fetuses were delivered by cesarean section. Before umbilical cord occlusion, the tracheal plug was removed by releasing the leading suture at its exit from the lamb’s neck, then applying traction to the other suture exiting the lamb’s mouth. After suctioning the upper airway, the lamb was intubated and the umbilical cord was clamped and divided. RESULTS

This two-trocar, video-fetoscopic technique was performed successfully in all four fetuses, with a sequential decrease in operating time (median, 3.5 hours; range, 5.5 to 2.3 hours). Although all fetuses were alive at the end of the procedure, only two survived through delivery. The first fetus to undergo video-fetoscopic tracheal occlusion was dead the following morning, probably as a result of prolonged exposure to halothane anesthesia. Another fetus was discovered stillborn in the cage at 136 days’ gestation, having been alive (by sonographic assessment) 1 week previously. At the time of necropsy, there was evidence of interval lung growth, with intact tracheal occlusion. The other two fetuses underwent cesarean delivery, and their tracheal plugs were removed; endotracheal intubation was achieved before umbilical cord division. After a period of positive pressure ventilation, the lambs were extubated and allowed to breathe spontaneously. Although neither lamb was capable of unsupported spontaneous ventilation, necropsy performed after euthanization showed enlarged lungs and reduced viscera, consistent with the physiological effects observed in our previous trachealocclusion animal experiments.

DISCUSSION

The combination of endotracheal and endoamniotic endoscopy performed through commercially available trocars adapted for uterine use facilitates a relatively simple technique of tracheal occlusion, which could be applied io human fetuses with CDH. The primary advantage of this technique over dpen methods of in utero therapy is that it avoids the lo- to 15-cm hysterotomy currently used to achieve fetal exposure, thereby theoretically reducing the risks of postoperative preterm labor. We fbund that the success of this video-fetoscopic procedure was dependent on attention to several important technical details. Endoscopic access followed by noncollapsible, secure trocar access to the fetal tracheal below the vocal cords appeared to be the most important step in achieving tracheal occlusion using our foam PLUG technique. We used a trocar with curvature and length that approximated those of the fetal lamb oropharynx; therefore, direct introduction into the lamb’s mouth typically placed the end of the trocar in the supraglottic region. Without the trocar in this position, videpscopic endotracheal intubation with the flexible bronchoscope was aImost impossible. Another important point is that videoscopic tracheal intubation must be carried out within fetal lung fluid. This fluid contains desquamated cells and other debris that cause significant light scatter and impair optical visualization. Attempts to insufflate the fetal airway with gas caused a continuous “bubbling” that further obscured the view. We found that continuous low-flow saline irrigation through the instrument channel of the bronchoscope yielded excellent supraglottic visualization and facilitated endoscopic intubation. Once the trachea had been intubated, it was relatively easy to advance the trocar over the bronchoscope, beyond the cords. With the trocar cuff inflated, not only was the fetal head internally immobilized, but there was secure 5-mm access to the trachea below the vocal cords. Endoamniotic visualization of the puncture-wire exit site on the lamb’s neck also was carried out through a fluid medium. In con&-ast to a previous fetoscopic surgical study in wliich carbon dioxide (CO,) insufflation was used,6 we found that visualization through amniotic fluid was adequate, provided the fluid was continuously exchanged for clear saline. Avoidance of CO2 insufflation within the uterus may be important, particularly in light of a recent fetal lamb study showing that intrauterine CO2 insufflation at pressures equal to resting amniotic pressures causes severe fetal hypercapnea and acidosis, presumably through passive CO2 diffusion into the fetus.’ The physical constraints of the small endoamniotic

1338

SKARSGARD

“working space,” visualization through fluid rather than air, and the potentially adverse effect of insufflating pressure on placental vascular resistance8 emphasize the limitations of video-fetoscopic surgery. Whereas standard laparoscopic procedures such as cholecystectomy adhere to the principles of surgical exposure through conventional dissection, endoamniotic procedures must be performed with a minimal number of small “ports” and with limited dissection. The Fetendo-PLUG procedure is a good example of a technique that conforms to these limitations by using only two small uterine ports, a minimal amount of endoamniotic instrumentation, and a technique of occlusion that relies on tracheal access from “within” rather than from outside. The efficacy of fetal tracheal ligation and tracheal occlusion in experimental models of CDH9 led to our recently reported experience with temporary tracheal occlusion in a 27-week human fetus with CDH.3 After delivery of the fetal head through a hysterotomy, the trachea was exposed through a transverse neck incision, and a guide wire was introduced into the trachea (through an B-gauge needle) and advanced retrograde through the larynx and out the mouth. The technique used to produce tracheal occlusion once guide-wire access to the trachea had been obtained was identical to that employed in these video-

ET AL

fetoscopic experiments. With ultrasound localization and a favorable fetal lie, trocar access to the fetal oral cavity followed by bronchoscopic guide-wire access to the trachea make tracheal occlusion without hysterotomy a potential therapeutic option for the human fetus with CDH. In addition to CDH, other fetal anomalies may be amenable to minimally invasive in utero therapy. Experimental models of hysteroscopic surgical repair currently exist for obstructive uropathy6 and for congenital cleft lip and palate.lO For the fetus that requires chronic intravascular access, an experimental hysteroscopic technique of placental vessel catheterization has been developed.” Human, hysteroscopic umbilical cord ligation of an acardiac twin with an associated twin-reversed arterial perfusion sequence (causing hydrops of the normal or “pump” twin) has been reported.“+13 The future of in utero therapy for antenatally diagnosed feta1 anomalies is dependent on the development of other “video-fetoscopic” techniques that require minimal intrauterine access and fetal manipulation, and thereby reduce postoperative fetal morbidity and mortality. As these techniques develop, new interventions for fetuses with correctable congenital anomalies are likely to evolve.

REFERENCES 1. Harrison MR, Adzick NS, Estes JM, et al: A prospective study of the outcome for fetuses with diaphragmatic hernia. JAMA 27:382-384, 1994 2. Longaker MT, Golbus MS, Filly RA, et al: Maternal outcome after open fetal surgery: A review of the first 17 human cases. JAMA 265:737-741,1991 3. Harrison MR, Adzick NS, Flake AW, et al: Response of the hypoplastic lung to tracheal occlusion in fetuses with diaphragmatic hernia. J Pediatr Surg (in press) 4. Harrison MR, Jester JA, Ross NA: Correction of congenital diaphragmatic hernia in utero I. the model: Intrathoracic balloon produces fatal pulmonary hypoplasia. Surgery 88:174-1821980 5. Beater JB, Skarsgard ED, Hedrick MH, et al: The “PLUG” odyssey: Adventures in experimental fetal tracheal occlusion. J Pediatr Surg 30:361-364, 1995 6. Estes JM, MacGitlivray TE, Hedrick MH, et al: Fetoscopic surgery for the treatment of congenital anomalies. J Pediatr Surg 27~950-954, 1992 7. Luks FI, Deprest J, Marcus M, et al: Carbon dioxide

pneumoamnios causes acidosis in fetal lamb. Fetal Diagn Ther 9:105-109, 1994 8. Skarsgard ED, Bealer JF, Meuli M, et al: Fetal endoscopic (“FETENDCY) Surgery: The relationship between insufflating pressure and the fetoplacental circulation. J Pediatr Surg 30:11651168,1995 9. Hedrick MH, Estes JM, Sullivan KM, et al: Plug the lung until it grows (PLUG): A new method to treat congenital diaphragmatic hernia in utero. J Pediatr Surg 29:612-617, 1994 10. Estes JM, Whitby DJ, Lorenz HP, et al: Endoscopic creation and repair of fetal cleft lip. Plast Reconstr Surg 90:750-756, 1992 11. Hedrick MH, Jennings RW, MacGillivray TE, et al: Chronic fetal vascular access: A new approach to the care of the fetal patient. Lancet 342:1086-1087,1993 12. Quintero RA, Reich H, Puder KS, et al: Brief report: Umbilical-cord ligation of an acardiac twin by fetoscopy at 19 weeks gestation. N Engl J Med 330:469-471,1994 13. McCurdy CM, Childers JM, Seeds JW: Ligation of the umbilical cord of n acardiac-acephalus twin with an endoscopic intrauterine technique. Obstet Gynecol82:708-711, 1993