Surg Endosc (1996) 10:820-824

Surgical Endoscopy © Springer Verlag New York Inc. 1996

Transesophageal echocardiography in fetal sheep A monitoring

tool for open and fetoscopic

cardiac procedures

T. Kohl, 1 E. J. Stelnicki, 2 K. J. VanderWall, 2 Z. Szabo, a E. Ko, 4 S. W. Bruch, 2 M. R. Harrison, 2 N. H. Silverman, s F. L. Hanley, 1 T. M. Chou 4 1Division of Pediatric Cardiothoracic Surgery, University of California, San Francisco, 505 Parnassus Avenue, San Francisco, CA 94143, USA 2 Fetal Treatment Center, University of California, San Francisco, 505 Parnassus Avenue, San Francisco, CA 94143, USA 3 Department of Surgery, University of California, San Francisco, 505 Parnassus Avenue, San Francisco, CA 94143, USA 4 Division of Cardiology, University of California, San Francisco, 505 Parnassus Avenue, San, Francisco, CA 94143, USA s Division of Pediatric Cardiology, University of California, San Francisco, 505 Parnassus Avenue, San Francisco, CA 94143, USA Received: 12 September 1995/Accepted: 18 December 1995

Abstract. Background: Cardiac procedures in exteriorized fetuses or assisted by fetoscopy require monitoring capabilities not attended by conventional maternal transabdominal echocardiography. Methods: We, therefore, assessed the potential of fetal transesophageal echocardiography (TEE) utilizing an intrav a s c u l a r ultrasound catheter (IVUC) for fetal cardiac monitoring. W e inserted a 10-F-10-MHz IVUC into the esophagus in 12 exteriorized fetal sheep and by a fetoscopic approach in 4 fetal sheep. Cardiac events were observed. Heart rate, cardiac rhythm, patency of the foramen ovale and ductus arteriosus, and the width of the branch pulmonary arteries could be assessed in all fetuses. Ventricular contractility could be assessed only in fetuses weighing less than 2.5 kg. Larger fetuses did not allow adequate imaging of the apical portion of the ventricles because of limited tissue penetration of the IVUC. Fetal TEE permitted placing small guide wires in the cardiac atria and left ventricle. Short-lived premature beats following intracardiac manipulations of these wires could be observed by fetal TEE in all cases. Results: At autopsy, no complications from IVUC insertion were observed in the exteriorized fetuses. Fetoscopic placement of the IVUC resulted in minor perioral skin erosion in two nonexteriorized fetuses. Conclusions: In conclusion, fetal TEE can be achieved with minor fetal injury and may provide useful information during open and fetoscopic cardiac procedures. Further improvements in IVUC design will permit the application of this technique to monitor human fetal cardiac procedures.

Key words: Fetal transesophageal echocardiography - - Intravascular ultrasound catheter - - Fetal cardiac intervention - - Fetal cardiac surgery --7 Fetoscopy

Correspondence to: T. Kohl

Fetal cardiac interventions have been performed on selected human fetuses with life-threatening cardiac anomalies. M o n i t o r e d by t r a n s a b d o m i n a l fetal e c h o c a r d i o g r a p h y , needle, guide wire, and interventional catheter were percutaneously introduced into the fetal heart to perform balloon valvuloplasties to treat severe aortic or pulmonary valve obstructions [1, 3, 9, 10]. In fetuses with complete heart block, the same monitoring technique was chosen to advance a pacing wire into the right ventricle [2, 13]. Transabdominal fetal echocardiographic monitoring, however, may interfere with surgical access or positioning of trocars and other devices during currently developed open [5-7] or fetoscopic cardiac procedures. Therefore, an alternative monitoring approach for the new procedures is desired. Since transesophageal echocardiography (TEE) has become valuable to monitor cardiac surgeries in children [11, 12] this approach may also be useful in the fetus for the same purpose. We, therefore, assessed the potential of fetal TEE by utilizing an intravascular ultrasound catheter (IVUC) for fetal cardiac monitoring in a fetal sheep model.

Materials and methods

Surgical preparation We studied a total of 16 time-dated fetal sheep (90-135 days; 145 = term) ranging from 0.5 to 4.3 kg in weight. In 12 of the 16 sheep, the ewe and her fetus were anesthetized with a maternal intravenous injection of ketamine hydrochloride (10-20 mg/kg). After positioning the ewe supine, the maternal abdominal wall was locally anesthetized with 10 ml of 1% lidocaine hydrochloride and the fetal head was exteriorized through a midline laparotomy and hysterotomy. In each of these exteriorized fetuses, a 10F-10-MHz IVUC was inserted via the mouth into the esophagus (Fig. 1A). In four fetuses, the IVUC was placed by a fetoscopic approach (Fig.

821

Fig. 1. 10-F-10-MHz intravascular ultrasound catheter inserted into the fetal esophagus in an exteriorized 90-day-gestation fetus (A) and during the fetoscopic procedure in a 110-day-gestation fetus (B). Motorddve+ electronic connection

+ 3

Fixed 10 MH; Transducer

Dlqve-Cable

r

~j~ Flush-Port

+

c

?

Angu[ated Mirror

Fig. 2. Schematic showing the percutaneous setup after insuffiation of the am~iotic sac. The first trocar (1) for insufflation and fetal visualization is placed by maternal transabdominal ultrasound. The second and third trocars are used for fetal forceps manipulation (2) and imaging catheter insertion (3). Fig. 3. Schematic of intravascular ultrasound catheter (IVUC). The catheterconsists of a fixed transducer and a rotating mirror assembly. The rotating mirror is angulated such that a 360 ° imaging plane is created perpendicular to the catheter long axis. The tissue penetration of the ultrasound beam is between 4 to 5 cm. The frame rate is 28 frames per second.

1B). For this approach, four ewes underwent general anesthesia with haiothane (0.5-2.0%) and were ventilated with 100% oxygen. The ewes were placed supine on a table in a reversed Trendelenburg position. In two of these ewes, a lower vertical midline laparotomy was performed to expose the uterus. The gravid horn was palpated to determine fetal position and three trocars (Dexide Inc., Fort Worth, TX) were placed through the uterine wall into the amniotic cavity. In the other two ewes, maternal transabdominal ultrasound (Sonos 1500, Hewlett-Packard, Sunnyvale, CA) was used to assess fetal lie and to safely place three trocars percutaneously in the amniotic cavity. This strategy obviated the need for matenlal laparotomy. After low-pressure air inflation of the uterus (2-3 mmHg), the fetus was visualized with a 5-mm 0 ° endoscope (Karl Storz Endoscopy America, Culver City, CA). The IVUC was then inserted into the fetal esophagus by forceps manipulation of the fetal head and mouth (Fig. 2).

Imaging device A commercially available intravascular ultrasound system with a 10-F IVUC (Boston Scientific Corporation, Sunnyvale, CA) was used for fetal TEE monitoring. The imaging catheter consists of a fixed 10-MHz transducer and a rotating mirror assembly (Fig. 3). The mirror is angulated 45 ° and is rotated by a cable extending through the catheter shaft, which is connected to a motor drive at the proximal end. The ultrasound beam reflects against the mirror and creates a two-dimensional 360 ° imaging plane perpendicular to the long axis of the catheter. Real-time images are obtained at 28 frames per second. The tissue penetration for this catheter is 4 to 5 cm, the axial resolution is 0.45 ram.

Protocol After insertion into the fetal mouth, the IVUC was advanced into the esophagus until the fetal heart was visualized in both exteriorized and

nonexteriorized fetal sheep. The screen image was rotated to display cardiac orientation in anatomic format. Heart rate, rhythm, patency of the foramen ovate, and cardiac contractions were assessed in a four-chamber view. By cranial withdrawal of the IVUC from the four-chamber view, the patency of the ductus arteriosus and the width of the branch pulmonary arteries were observed. To obtain a fetal ECG during three fetal TEE studies, two circular electrodes were attached 2 cm apart to the IVUC via a heat-shrink membrane. The ECG wires were connected to a cardiac monitor that was interfaced with the intravascular ultrasound system. In two other exteriorized fetuses, the effect of indomethacin infusion on the ductus arteriosus was observed. In three exteriorized fetuses, a 0.014-inch guide wire was advanced via the umbilical vessels and TEE-assisted guidance of the wire into the fetal heart was attempted. All studies were recorded continuously on super-VHS tape. Following each study, both ewe and fetus were euthanatized with overdose pentobarbital sodium. Procedure-related complications were evaluated at autopsy. The study protocol was approved by the local Committee on Animal Research and performed in accordance with the institutional guidelines on research animal use.

Results T h e I V U C c o u l d b e i n s e r t e d i n t o t h e e s o p h a g u s i n all 12 exteriorized fetal sheep and by endoscopic fetal manipulat i o n i n all f o u r n o n e x t e r i o r i z e d f e t a l s h e e p . I n t h e l a t t e r group, the time interval needed for manipulation of fetal head position, insertion of the third trocar, and placement of the IVUC was less than 8 min.

822

Fig. 5. Fetal TEE monitoring of (A) the pulmonary artery bifurcation

and (B) the ductus arteriosus (DA) in two 90-day-gestation fetal sheeps. Changes in the diameter of these vessels may be induced by matenaal anesthesia and procedure-related cardiovascular manipulations. The ECG in A was obtained by mounting the intravascular ultrasound catheter (IVUC) with a unipolar ECG lead. MPA = main pulmonary artery, AoA = aortic arch, AoD = descending aorta. A = anterior, P = posterior, L left, R = right.

Heart rate and cardiac rhythm could be monitored continuously in all fetuses (16/16 fetuses; 100%). In three fetuses in whom intracardiac guide wire manipulations were performed, short periods of premature contractions were observed that cleared with wire repositioning. In fetuses weighing less than 2.5 kg, heart rate and rhythm were best observed at a four-chamber view (Fig. 4A, B). Heavier fetuses (8/16; 50%) were too large to allow adequate imaging of the ventricles. In those, heart rate and rhythm were best assessed from the rhythmic distensions of the great arteries with ventricular systole. In three exteriorized fetuses, a fetal E C G was obtained, providing complementary information about cardiac rhythm and timing of cardiac events (Fig. 5). The patency of the foramen ovale could be directly observed in 12/16 fetuses (75%) and was also indicated by the free motility of the septum primum on the left atrial side throughout the cardiac cycle. Because of the limited penetration of the ultrasound beam (4-5 cm), ventricular contractility could be assessed from a four-chamber view in only those fetuses weighing less than 2.5 kg (8/16; 50%).

The width of the branch pulmonary arteries could be observed in 13/16 fetuses (81%) (Fig. 5A). In the remaining three fetuses, air within the trachea and mainstem bronchi prevented satisfactory imaging of the pulmonary arteries. The ductus arteriosus was imaged throughout its entire course between main pulmonary artery and descending aorta in all fetuses regardless of gestational age (Fig. 5B). In two exteriorized fetuses (125-130-day gestation), indomethacin-induced subtotal constriction of the ductus arteriosus was observed by fetal TEE and confirmed at autopsy (Fig. 6). TEE-assisted placing of a 0.014-inch guide wire into the fetal heart was performed in three small 90-100day-old exteriorized fetuses weighing between 550 to 700 g. In two of these fetuses, the wire was introduced into the umbilical vein and could be guided by fetal TEE repeatedly from the ductus venosus-inferior caval vein junction into the right atrium and right ventricle as well as across the foramen ovale into the left atrium. In the third fetus, the wire was introduced via the umbilical artery and placed across the aortic arch into the left ventricle.

Fig. 4. Fetal TEE four-chamber plane. A Systolic frame. B Diastolic frame. At this view heart rate, rhythm and cardiac contractions can be best assessed. LV = left ventricle, R V = right ventricle, LA = left atrium, RA = right atrium. A = anterior, P = posterior, L = left, R = right. 1VUC = intravascular ultrasound catheter.

(PAB)

823

Fig. 6. Fetal TEE monitoring before (A) and after (B) indomethacin-induced constriction of the ductus arteriosus (DA) in 125-day old fetal sheep. The asterisks in A indicate the area of unobstructed connectionbetween ductus and descending aorta (AoD) before Indomethacin administration. MPA = main pulmonaryartery, RPA = right pulmonary artery. A = anterior, P - posterior, L = left, R = right. IVUC = intravascular ultrasound catheter. At autopsy, no oropharyngeal or intraesophageal lesions were observed in the exteriorized fetuses. In two of the four nonexteriorized fetuses, fetoscopic IVUC placement into the esophagus resulted in minor perioral skin erosion due to forceps manipulation of fetal head position. No maternal or fetal complications from low-pressure uterine inflation with air were observed.

Discussion This study in fetal sheep shows that fetal TEE permits fetal cardiac monitoring in exteriorized and nonexteriorized fetuses. This approach has the potential to provide useful information about cardiac events during currently developed open and endoscopic fetal cardiac procedures. During open fetal heart surgery, fetal TEE may be as valuable in defining cardiac function and results of repair as it has been in pediatric populations [11, 12]. During fetal cardiac catheterizations to treat severe semilunar valvar stenosis, fetal TEE might be employed to guide interventional catheters into the fetal heart. In our study, we were able to reliably place 0.014-inch guide wires into various cardiac chambers in several exteriorized fetuses. As a prerequisite for fetoscopically assisted cardiac catheterizations, intraesophageal IVUC placement by a fetoscopic approach could be achieved in four of our study fetuses with only minor perioral fetal injury. The next logical step in the effort to perform fetoscopic cardiac catheterization requires the definition of a safe and reliable method for fetal cardiovascular access. Fetal TEE permits continuous real-time monitoring of fetal heart rate, cardiac rhythm, and contractions. Furthermore, the patency of the foramen ovale and ductus arteriosus as well as branch pulmonary artery width can be monitored to assess redistribution of fetal blood flow due to maternal anesthesia or procedure-related manipulations. Indomethacin-induced constriction of the arterial duct could clearly be defined in our study. Description of fetal cardiac rhythm can be further improved by mounting the IVUC with an ECG lead. This method produced reliable ECG tracing and provided detailed information about cardiac rhythm and timing of cardiac events in three of the study fetuses.

Conventional transabdominal or transvaginal fetal imaging techniques can provide information similar to fetal TEE but they are more dependent on fetal position and size. In addition, they may also interfere with the surgical access during open and fetoscopic cardiac procedures. Fetoscopic cardiac procedures in a gas-distended amniotic cavity [4] require imaging capabilities not attended by transabdominal echocardiography. Although uterine gas distension provides unparalleled fetal visualization and exposure for intraesophageal IVUC insertion, the safety of this approach needs to be defined before it can be applied to human fetoscopic cardiac procedures.

Current technical limitations o f f e t a l T E E

Fetal TEE monitoring might be safely performed for a few hours following the intervention but continuous postprocedural monitoring would be restrained mainly for two reasons. First, the likelihood of amnionitis increases if the catheter is left in utero for prolonged periods. Second, the catheter lifetime is limited to 2-4 h of continuous use. Fetal monitoring after open fetal cardiac procedures may be continued by a radiotelemetric device which has been successfully used during noncardiac surgical procedures in human fetuses [8]. This device is implanted subcutaneously in the fetal axilla at the time of repair and continuously records fetal ECG, temperature, and intraamniotic pressure. A pulse oxymeter can be placed around a fetal hand during open surgical procedures to measure fetal oxygen saturation [7] and would complement to fetal TEE monitoring. If fetal TEE is planned after gaseous insufflation of the amniotic cavity to facilitate fetal cardiovascular access, in utero positioning of the fetus should be achieved in such a way that a sufficient amount of lung fluid is retained in the fetal airways because gas in these structures would interfere with satisfactory cardiac monitoring from the intraesophageal site. The tissue penetration of the 10-MHz IVUC used in this study was limited to 4-5 cm. Therefore, in study fetuses weighing more than 2.5 kg, the visibility of the apical portion of the ventricles was impaired. While information

824 about heart rate and r h y t h m as well as ductus arteriosus and f o r a m e n ovale patency could still be obtained, cardiac contractions could not be assessed in these large fetuses since the apical portion of the ventricles was not visualized. Because fetal cardiac interventions will most likely be performed in smaller fetuses, this limitation seems to be only of academic interest. In our study, the fetuses were sacrificed i m m e d i a t e l y after the procedure. Thus no conclusions for the long-term outcome after fetal T E E could be drawn. In addition, the n u m b e r of fetal T E E studies done fetoscopically was to small to conclude net complications to the procedure. W e believe, however, that the ultimate fetal outcome will depend on the c o m b i n e d response of fetus, placenta, and uterus to fetal cardiovascular access and procedure. However, the intraesophageal presence of a m o n i t o r i n g tool that could be placed with m i n i m a l fetal and uterine injury m a y have little bearing on fetal outcome.

Conclusion Fetal transesophageal echocardiography can be achieved with m i n o r fetal injury and provides useful information during open and fetoscopic cardiac procedures. Further m i n i a turization in I V U C size in concert with increased tissue penetration w o u l d permit fetal T E E at all gestational ages. Phased-array technology and pulsed Doppler capabilities are desirable to further improve u p o n i m a g i n g quality and allow for additional information on fetal h e m o d y n a m i c s .

Acknowledgments. We thank Mrs. Christine Roman, Mr. Mario Tmjillo, and Mr. Vincente G. Lapiz for expert technical assistance. We acknowledge the support of Boston Scientific Corporation (Sunnyvale, CA), Karl Storz Endoscopy America (Culver City, CA), and Dexide, Inc. (Fort Worth, TX) for this study.

Dr. Kohl is supported by an educational grant (Ko 1484 / l-i) of the German Research Society (DFG)- Germany. Dr. Chou is funded by NIH grant F32 HL-090969-01.

References 1. Allan LD, Maxwell DJ, Carminati M, Tynan MJ (1995) Survival after fetal aortic balloon valvoplasty. Ultrasound Obstet Gynecol 5(2): 9091 2. Carpenter R, Jr, Strasburger JF, Garson A Jr, Smith RT, Deter RL, Engelhardt H Jr. Fetal ventricular pacing for hydrops secondary to complete atrioventricular block. J Am Coll Cardiol 8(6): 1434-1436 3. Crombleholme TIVl (1994) Invasive fetal therapy: current status and future directions. Semin Perinatol 18(4): 385-397 4. Estes JM, Szabo Z, Harrison MR (i992) Techniques for in utero endoscopic surgery. A new approach for fetal intervention. Surg Endosc 6(5): 215-218 5. Hanley FL (1994) Fetal cardiac surgery. Adv Card Sorg 5:47-74 6. Hanley FL (1994) Fetal cardiac surgery: steady progress in the laboratory [editorial]. Ann Thorac Surg 57(2): 279-280 7. Harrison MR (1993) Fetal surgery. West J Med 159(3): 341-349 8. Jennings RW, Adzick NS, Longaker MT, Lorenz HP, Harrison MR (1993) Radioteiemetric fetal monitoring during and after open fetal operation. Surg Gynecol Obstet 176(1): 59-64 9. Lopes LM, Cha SC, Kajita LK, Aiello VD, Jatene A, Zugaib M (1996) Balloon dilatation in the fetus--a case report. Fetal Diagn Ther (in press) 10. Maxwell D, Allan L, Tynan MJ (1991) Balloon dilatation of the aortic valve in the fetus: a report of two cases. Br Heart J 65(5): 256-258 11. Muhiudeen IA, Roberson DA, Silverman NH, Haas GS, Turley K, Cahalan MK (1992) Intraoperative echocardiography for evaluation of congenital heart defects in infants and children [see comments]. Anesthesiology 76(2): 165-172 12. Stevenson JG, Sorensen GK, Gartman DM, Hall DG, RiUenhouse EA (1993) Transesophageal echocardiography during repair of congenital cardiac defects: identification of residual problems necessitating reoperation. J Am Soc Echocardiogr 6(4): 356-365 13. Walkmshaw SA, Welch CR, McCormack J, Walsh K (1994) In utero pacing for fetal congenital heart block. Fetal Diagn Ther 9(3): 183-185