- Email: [email protected]
Correction of congenital diaphragmatic hernia in utero: VI. hard-earned lessons
Correction of Congenital D i a p h r a g m a t i c Hernia In Utero: VI. H a r d - E a r n e d L e s s o n s By Michael R. Harrison, N. Scott Adzick, Alan W. Flake, Russell W. Jennings, James M. Estes, Thomas E. MacGillivray, Jane T. Chueh, James D. Goldberg, Roy A. Filly, Ruth B. Goldstein, Mark A. Rosen, Charles Cauidwell, Anne H. Levine, and Lori J. Howell San Francisco, California 9 Extensive experimental work suggests that repair of congenital diaphragmatic hernia (CDH) in utero may salvage severely affected fetuses who otherwise have a high expected mortality despite optimal postnatal care including extracorporeal membrane oxygenation (ECMO). We have reported that repair of CDH in utero is physiologically sound and safe for the mother, but technically difficult especially when the liver is herniated into the fetal chest. In the 3 years since our last report (1989 to 1991), 61 additional patients were referred for consideration of in utero repair. Fetal repair was attempted in 14 with severe isolated left CDH diagnosed before 24 weeks gestation. Five fetuses died intraoperatively, from technical problems related to reduction of incarcerated liver and uterine contractions--problems which have subsequently been surmounted. Nine patients were successfully repaired. Four babies survived, two delivered prematurely and died, and three died in utero within 48 hours of repair. Intraoperative technical problems have been overcome; the factors limiting successful outcome are postoperarive physiologic management of the maternal-fetal unit and effective tocolysis to control preterm labor. Copyright 9 1993 by W.B. Saunders Company INDEX WORDS: Congenital diaphragmatic hernia, prenatal diagnosis, fetal surgery.
O N G E N I T A L diaphragmatic hernia (CDH) remains a frustrating clinical problem. Although less severely affected babies survive with modern postnatal surgical care including extracorporeal membrane oxygenation (ECMO), the majority of babies who are severely affected die despite all intervention. 1-n Because retrospective estimates of mortality for C D H vary widely and are flawed by a "hidden mortality" of unknown magnitude, we did a prospective study which shows the mortality for fetuses with correctable isolated diaphragmatic hernias diagnosed before 24 weeks is 60% despite optimal postnatal care including E C M O (submitted for publication). Salvage of these severely affected babies with isolated diaphragmatic hernias is an unsolved problem. The pulmonary hypoplasia that limits survival is an underdevelopment of both the parenchyma and the pulmonary vascular bed, 12-19and appears to be reversible both experimentally and clinically. 13,15,21,22However, many weeks or even months are required. E C M O is limited to 1 to 2 weeks, and while useful for marginal babies, it clearly cannot salvage severely affected babies. Long-term support or replacement of lung function after birth will require either an artifi-
C
Journal of Pediatric Surgery, Vo128, No 10 (October), 1993: pp 1411-1418
cial placenta or neonatal lung transplantation. Repair before birth with continued support on the "placental E C M O " while the fetal lung grows and recovers would be ideal. However repair in utero with continued gestation has proven to be a formidable challenge.16, 23 Because maternal safety is an overriding issue for fetal surgery, we first demonstrated in the nonhuman primate 24-26 and then in our first 17 patients 16,17 that hysterotomy and fetal surgery did not adversely affect the mother or her ability to have subsequent normal children. Our initial attempts to repair six fetuses with diaphragmatic hernia showed that although the repair was feasible, there were very significant technical difficulties particularly from herniated liver. 16 We now report that further experimental and clinical studies over the last 3 years have resolved many of the intraoperative technical problems 2s,29 and allowed a few successful repairs. However, most attempts were not successful, and this frustrating and disappointing experience has taught us many hard-earned lessons about deficiencies in our intraoperative and postoperative management, particularly our inability to effectively monitor and treat preterm labor. MATERIALS AND METHODS
Sixty-one patients were referred for consideration of in utero repair of prenatally diagnosed diaphragmatic hernia between January 1989and December 1991.Fetal repair was attempted in 14 with severe isolated lesions diagnosed before 24 weeks' gestation. All had normal chromosomes by amniocentesis or percutaneous umbilical blood sampling, had left-sided hernias with dilated stomach incarcerated in the chest and significant contralateral mediastinal shift. Only 4 had polyhydramnios.Although polyhydramnios correlates with poor outcome, it frequently appears later than the 22 to 26 weeks when these fetuses were evaluated. In all cases the familywas interviewedand counseled by a perinatologist,
From the Fetal Treatment Center and the Departments of Surgery, Obstetrics, Gynecology, Reproductive Sciences, Anesthesiology, Radiology, and Nursing, University of California, San Francisco, CA. Presented at the 1992 Annual Meeting of the Section on Surgery of the American Academy of Pediatrics, San Francisco, California, October 9-11, 1992. Address reprint requests to Michael R. Harrison, MD, Fetal Treatment Center, University of California, San Francisco, 513 Parnassus Ave, Room 585-HSE, San Francisco, CA 94143-0570. Copyright 9 1993 by W.B. Saunders Company 0022-3468/93/2810-0037503.00/0 1411
1412
HARRISON ET AL
Table 1. Fetal Diaphragmatic Hernia Repair: Analysis of Cases Case No. Survived
1
4
6
12
Died at operation
2
5
8
10
11
Repaired, died in utero
3
Clinical Presentation
Procedure
Course
27 yo G2P1; 24.5 wk Posterior fundal hysterotomy; subcostal exmale; liver: lateral posure, diaphragm patch, abdominal segment up; plapatch; HR 120-150; 54 min centa: right anterior
Preterm labor ~ PROM; C/S at 32 wk; 1,920 g, Apgar 7 and 9; resp support 1 mo; fundoplication at 4 mo Posterior fundal hysterotomy; Subcostal Preterm labor ~ PROM; 27 yo G3P2; 25 wk feexposure; diaphragm patch, lined abC/S at 32 wk; 1,850 g, male; liver: down; Apgar 7 and 9; Rasp placenta: midline dominal patch; stable HR 130-150; 76 rain anterior support 2 da Posterior fundal hysterotomy; difficult fetal Pulmonary edema = toco30 yo G2P0; 24 wk feposition --~ bradycardia; diaphragm patch, male; liver: lateral lyrics; oligohydramnios --~ repair AF leak segment up; plaabdominal patch; HR >20; 61 min POD11; C/S at 28 wk centa: midline ante1,400 g. Apgar 3 and 8; rior rasp support 1 mo; fundoplication and repair at 2 mo 19 yo G3P1; 25.5 wk Transverse fundal hysterotomy; double nu- Benign postoperative male; liver: left lobe chal cord; "two-step" exposure: thoracic course; oligohydramup; placenta: low and abdominal exposure; diaphragm nios; C/S at 34 wk; anterior patch; abdominal patch; HR > 130; 66 2,460 g, Apgar 7 and 7; min resp support 3 wk; fundoplication at 1 mo Transverse fundal hysterotomy; difficult fe33 yo G3P2; 25 wk male; liver: left lobe tal position --* brady with manipulation; difficult to reduce incarcerated liver; braup; placenta: middys <80 x 10; unresponsive to resuscitaline anterior tion; 102 min 37 yo G3P1; 24 wk fe- Anterior hysterotomy; subcostal exposure male; liver: lateral difficult ~ subcapsular hematoma -* segment up; platransfusion; diaphragm patch; bradycarcenta: posterior dia with abdominal patch; cord puncture lateral and transfusion --~ spasm; unresponsive to resuscitation; 78 min 19 yo G2P0; 23 wk fe- Left paramedian hysterotomy; liver reducmale; liver: minimal tion ~ bradycardia ~ unresponsive to liver; placenta: latresuscitation; 85 rain eral right 33 yo G4P~; 25wk Anterior, low dght hysterotomy; velamenmale; liver: left lobe tous insertion of cord; large liver mass could not be reduced; liver resection with in chest; placenta: left fundal and anteargon beam --> sinus venosus bleed; resuscitation attempted via puncture of rior velamentous cord --> unresponsive; 100 min Anterior midline hysterotomy; Broviac cen32 yo G2P0; 24.5 wk male; liver: left lobe tral access; bradycardia from beginning ~ uterine contractions throughout; resusciup; placenta: posterior tation w/crystalloid, HCO3; liver ligated at "waist" and resected with no blood loss; uterine contractions/cord compromise --~ fetal death; 83 min
Comment Postnatal morbidity preterm delivery, need better tocolytics
Postnatal morbidity preterm delivery, need better tocolytics AF leak diagnosed by amnioinfusion and repaired, oligohydramnios 1' morbidity after birth
"Two-step" exposure excellent allowed liver reduction, need more time in utero to ~ morbidity
Autopsy: repair of diaphragm OK; kinked umbilical vein from liver reduction
Subcostal exposure inadequate "waist" in liver ---could have resected
Autopsy: "battledore" placenta, repair OK
Abdominal closure tight; cord puncture led to cord spasm
Autopsy: repair OK, .7 um- Reduction of liver ~ brabilical vein kinked dycardia (need to resect) Autopsy: sinus venosus distorted to left; cut during resection
Subcostal exposure inadequate, liver resection possible, need to locate sinus venosus by Doppler before dividing "waist"
Autopsy: repair OK, liver resection OK, no significant bleeding
"Two-step" exposure excellent, liver resection without problem, all foiled by uterine contractions, need to reassess anesthesia/tocolysis
Posterior fundal hysterotomy; core wrapped Fetus stable HR >120, 24 yo G1P0; 25 wk male; liver: left lobe around neck ~ difficult exposure; subgood anatomy by costal exposure difficult to retract liver; sonogram postoperaup anterior; plabrady <80 x 12; diaphragm patch, abtive; fetal demise 1st centa: right anterior dominal patch - ? tight; 54 min postop night; autopsy: repair OK, liver ischemia
Postoperative monitoring by tocodynamometer inadequate, ? late death umbilical distortion
DIAPHRAGMATIC HERNIA IN UTERO
1413
Table 1. Fetal Diaphragmatic Hernia Repair: Analysis of Cases (Cont'd) Case
No.
Clinical Presentation
Procedure
Course
Comment
9
25 yo G2P1; 28 wk fe- Anterior oblique hysterotomy; subcostal male; liver: left lobe exposure inadequate for liver; brady up; placenta: left <80 • 20 when liver reduced; diaphragm posterior patch, abdominal patch; 99 min
Fetus stable with good anatomy; 14 h postoperation developed bradycardia --> died; autopsy: repair intact, unexplained demise
13
39 yo G4P0; 25.5 wk Posterior fundal hysterotomy; "two-step" male; liver: left lobe exposure: excellent; lateral segment left up, good waist; plalobe resected without problems; diacenta: anterior midphragm patch; abdominal silo; fetal HR line stable > 140 throughout; uterine contractions problem; 69 rain
Stable fetus--good HR; Fetal liver resection feasi30 h postoperation deble, postoperative manveloped bradycardia; agement inadequate, unresponsive to atfetal monitor OK, monitempted resuscitation tor of labor poor, tocoincl PUBS; autopsy: lytics (MgSO4 + terburepair intact, no bleedtaline) toxic, need ing, Hct 43 maternal/fetal ICU AF volume remained low; Tocolysis inadequate vaginal leak and PROM contractions --> PROM, at 26 wk; C/S at 31 wk; oligohydramnios --* 1,425 g, Apgars 3 and lung hypopiasia 6; died post 2 h; autopsy: pulmonary hypoplasia, small disruption diaphragm Difficult postoperative Intraoperative problems course due to toxicity solved; postoperative of tocolytics; preterm management inadlabor poorly controlled; equate; tocolysis AF vaginal leak --, (MgSO4 + terbutaline) PROM at 28 wk; C/S at toxic; need new anes31 wk; 1,565 g, Apgars thetic/tocolytic man3 and 7; died post 2 wk agement + fetal ICU
Died after birth
29 yo G3P2; 24 wk male; liver: no liver up; placenta: posterior midline
14
Anterior transverse hysterotomy; difficult uterus tight + herniated cord; diaphragm and abdominal patch; HR stable > 120 throughout; 63 min
43 yo GzP0; 25.5 wk Low anterior, oblique hysterotomy; twomale; liver: left lobe step exposure; reduced liver with diffiup; placenta: anteculty; diaphragm patch, abdominal silo; rior left, fundal bradycardia <90 • 3 w/liver manipulation; HR stable > 120 after liver down; PUBS Hct 46, ABG OK; 79 min
Difficult but successful repair, radiotelemeter fetal ECG excellent, angiogram -~ distorted umbilical vein ? = late death
Abbreviations: G, gravida; P, para; HR, heart rate; C/S, cesarean section; Resp, respiratory; POD, postoperative day.
surgeon, counselor, anesthesiologist, clinical nurse specialist, and psychiatric social worker. Families were offered intervention only when all parties agreed that it was appropriate medically and psychologically for that family. Families signed a consent form approved by the Committee on Human Research at UCSF. Financial considerations were frequently limiting. Many families could not afford to come to San Francisco for evaluation. Although 12 of 14 had some third-party coverage, only five carriers agreed to pay for even part of the fetal procedure. Two paid in full, and the others covered 33% to 80% of hospital costs. The unreimbursed portion of the hospital bill was borne by the UCSF Medical Center and the professionals involved. More than 20 professionals donated their services through the UCSF Fetal Treatment Center. More recently, American Airlines has provided air travel though its "Miles for Kids" Program and the March of Dimes has helped the families with local accommodations. RESULTS
The clinical course and outcome of each case is presented in Table 1. The "hard-earned" lessons derived from this experience are summarized below.
Intraoperative Death (Cases 2, 5, 8, 10, and 11) Five fetuses died intraoperatively from technical problems related to reduction of incarcerated liver,
providing the following hard-earned lessons. (1) Fetal monitoring by standard electrocardiogram (ECG) leads or pulse oximeter was inadequate. The radiotelemeter ECG 3~ solved this problem. (2) The standard subcostal incision was inadequate to deal with a herniated liver. The "CDH two-step ''28 solved this problem. (3) An incarcerated liver could not be reduced without kinking the umbilical vein. Experimental work on the course of the umbilical vein 31 and on new techniques for fetal liver resection (submitted for publication) solved this problem. (4) The ductus venosus is distorted to the left by herniation and is easily injured during resection. Intraoperative Doppler to identify the course of the ductus venosus solved this problem. (5) Finally, even when all these technical maneuvers were accomplished (case 11), success was spoiled by uncontrolled uterine contractions which began as we made the hysterotomy, caused repeated bradycardias throughout the procedure, and did not respond to deepending isofluorane anesthesia. Although halogenated agents appeared to be an effective tocolytic based on our original
1414
primate work, 24-26 it was becoming clear that a new approach to both anesthesia and tocolysis was needed.
Postoperative Fetal Death (Cases 3, 9, and 13) Three had successful repairs only to die in the perioperative period. The hard-earned lessons from these frustrating cases led us to the concept of postoperative maternal-fetal intensive care. (1) Periodic fetal heart rate testing by Doppler was inadequate after fetal surgery. Bradycardia can progress quickly to fetal demise. The radiotelemeter ECG 3~ solved the postoperative as well as the intraoperative monitoring problem. (2) Monitoring of uterine activity by external tocodynamometer was inadequate to allow adjustment of tocolytics. This problem has been solved experimentally by using the radiotelemeter to monitor uterine electromyeogram and intrauterine pressure. (3) Even when fetal deterioration was accurately detected (case 13), attempted fetal resuscitation was hindered by our inability to access the fetal circulation. Recently, we have developed endoscopic catheterization of placental vessels for chronic fetal vascular access, a9 (4) Finally, postoperative tocolysis with magnesium sulfate, terbutaline, and indocin proved inadequate and even toxic.
Neonatal Death (Cases 7 and 14) After successful fetal repair, preterm labor and premature rupture of membranes led to preterm delivery and neonatal death in two cases. We learned several lessons. (1) Premature rupture of membranes and persistent oligohydramnios prevented lung growth and contributed to premature delivery and subsequent neonatal death. We then modified our technique of closing the uterus to include removal of staples, use of stays, and use of fibrin glue to help prevent amniotic fluid from dissecting under the membranes and causing a vaginal leak. (2) The standard obstetrical regimen for tocolysis was inadequate and even toxic. Case 14 had a very difficult postoperative course due to poorly controlled labor and the toxicity of magnesium sulfate and terbutaline. For the first time the patient had arterial and central venous lines placed intraoperatively and was monitored in the intensive care unit for the first 3 days. Maternal volume depletion and its adverse effects on 9uterine blood flow were documented by Doppler ultrasonography. Even after discharge from the hospital, preterm labor was poorly controlled and premature rupture of membranes was documented by a vaginal leak at 28 weeks. This case illustrated the need for improved postoperative intensive care and led us to challenge the standard obstetrical regimen for managing preterm labor with magnesium sulfate,
HARRISON ET AL
terbutaline and maternal fluid restriction. A new approach to management of the fetal-placentalmaternal unit after hysterotomy and fetal surgery would be based on reexamination of maternal-fetal anesthesia, monitoring, and tocolysis in experimental models. 29-31
Survivors (Cases 1, 4, 6, and 12) Four babies survived. Even the successful cases provided some lessons. (1) Cases 1 (reported previously23) and 4 delivered too early (32 weeks) and experienced significant postnatal morbidity which would have been minimized if pregnancy was prolonged for at least another month. (2) Amniotic fluid leak with oligohydramnios can sometimes be corrected. In case 6, the fetus did well but mother had an extremely difficult time with the tocolytics magnesium sulfate and terbutaline which led to mild pulmonary edema. Postoperative oligohydramnios was initially thought to be due to indomethacin, but an amnioinfusion showed a leak through the hysterotomy site. On postoperative day 11, the patient was reexplored and an amniotic fluid leak through a small uterine disruption was repaired. The hysterotomy wound edges looked ischemic probably secondary to the occlusive clamps used on the uterine incision at that time. This problem was solved by a special nonocclusive backbiting clamp-retractor that provides exposure and hemostasis without pressure necrosis of the uterine edges. For closure, we now remove the staples to allow approximation of the full-thickness of uterine muscle by interrupted stay sutures in addition to two layers of running Maxon, and fibrin glue to seal the inner layer. (3) Inability to control uterine contractions may have contributed to the uterine disruption and amniotic fluid leak as well as preterm labor. (4) Large diaphragmatic defects which require a prosthetic patch, whether repaired before or after birth, will often be complicated by later patch disruption with growth as well as gastroesophageal reflux. These problems once recognized can be corrected surgically as was done in cases 6 and 12. DISCUSSION
Diaphragmatic hernia is still an unsolved problem. The pathophysiologic rationale for in utero repair is compelling, 1-15 but repair of diaphragmatic hernia before birth has proven difficult. 16,23A succession of intraoperative technical problems have been slowly surmounted, only to encounter more challenges in postoperative care. The problem of accurate preoperative assessment of the extent of liver herniation has finally yielded to
DIAPHRAGMATIC HERNIA IN UTERO
careful evaluation of the umbilical vein by color Doppler. When the umbilical vein is deviated far to the left, the liver is usually up in the left chest. If the course of the sinus venosus remains below where the diaphragm should be, the herniated liver is amenable to reduction. If it appears well up in the chest, the entire liver is torqued into the chest and probably cannot be reduced without compromising umbilical flow. This is certainly true in the unusual right-sided diaphragmatic hernia as we have demonstrated in a recent case by showing that reduction of the torqued liver kinked not only the sinus venosus but also the inferior vena cava. In our experience, most fetuses with left diaphragmatic hernias have some liver herniated into the chest. This comes as a surprise, since pediatric surgeons seldom see liver in a CDH patient after birth. This is because babies who make it to surgery have already had the liver reduced from the chest by positive-pressure ventilation. Reduction of the liver after birth either by the respirator or by the surgeon causes little problems since there is no blood flow through the umbilical vein. Reduction of the liver before birth is devastating because it compromises the umbilical circulation. Herniated liver accounted for most of our failures. Disturbance of the umbilical circulation during or after liver reduction is implicated in all five intraoperative deaths and in the three early postoperative deaths. We have recently demonstrated using postmortem angiograms and vascular casts of the umbilical and hepatic circulations that kinking of the extrahepatic umbilical vessels can be prevented by dividing the falciform ligament that attaches the umbilical vein to the underside of the abdominal wall.31 In addition, we went back to the laboratory and developed techniques for fetal liver resection (submitted for publication). Now if the liver cannot be reduced without compromising umbilical circulation, the left lobe of the herniated fetal liver can be amputated as demonstrated in cases 11 and 12. The key consideration in amputating the liver is the position of the ductus venosus (see case 10) and this can now be delineated intraoperatively by Doppler ultrasound. The toughest lesson was that the subcostal incision that works so well after birth provides inadequate exposure to deal with the severe defects encountered in those fetuses. The solution was a carefully planned two-step approach using both a thoracotomy and a subcostal incision. This method allows reduction of viscera using a "push-pull" technique, reduction or amputation of incarcerated liver, and reconstruction of the diaphragm with a prosthetic patch. This technique is described in detail elsewhere. 2s
1415
Problems with uterine disruption and amniotic fluid leak (cases 6, 7, and 14) led to revision of our technique of opening and closing the gravid uterus. We have tried many ways to control bleeding from the cut uterus. Occlusive clamps made the edge of the uterus ischemic, and led to the development of a gentle back-biting, nonocclusive retractor that controis bleeding and provides exposure without causing ischemia. The stapling device we developed a decade ago prevents bleeding and seals the membranes to the myometrium. But we have now learned to remove the staples from the wound edge just prior to closure to allow muscle-to-muscle approximation. All layer stay sutures, fibrin glue, and a two-layer closure help prevent dissection of amniotic fluid resulting in vaginal leak. Even with these advances, complete repair of fetal CDH remains challenging. We have long sought another approach to improving fetal lung development that did not require the extensive manipulation necessary for total repair. The simple expedient of creating an abdominal wall defect and leaving the viscera externalized has proven unsatisfactory: decompression of the chest is inadequate and damage to the viscera can be significant, as demonstrated in one of our early cases. 16 Improving pulmonary surfactant production may be helpful for maturation33 but will not affect growth and development. Another less invasive approach appears promising. We have shown that impeding the normal egress of fetal lung fluid by controlled tracheal obstruction enlarges the hypoplastic lung and pushes the viscera back into the abdomen in fetal lambs. This new procedure, which we call "PLUG" (Plug Lung Until it Grows), may prove useful when total correction in utero is not possible. Significant obstacles to successful repair of diaphragmatic hernia and other life-threatening fetal lesions remain. The limiting factor is now anesthetic and tocolytic management. Based on our initial primate studies, 24-26 we thought that deep anesthesia with a halogenated agent (halothane, isofluorane) provided adequate anesthesia for fetus and mother as well as complete uterine relaxation. But our clinical experience has been disappointing. Indeed intraoperative contractions caused significant problems in at least half of these cases, and clearly ruined an otherwise successful repair in two (cases 1l and 13). In addition, fetal lamb studies have shown that halothane anesthesia depresses myocardial function in both the fetus and the mother and does not block the stress response that leads to placental hypoperfusion. 34We are investigating the possibility of blocking the fetal and maternal stress response with opioids. The most glaring deficiency in management is our
1416
HARRISON ET AL
inability to m a n a g e t h e p r e t e r m l a b o r i n d u c e d by h y s t e r o t o m y a n d fetal surgery. W e have b e e n f o r c e d to rely on t h e r e g i m e n u s e d ( o f t e n ineffectively) for s p o n t a n e o u s p r e t e r m labor. T h i s i n c l u d e s e x t e r n a l m o n i t o r i n g w i t h a t o c o d y n a m o m e t e r , b e d r e s t , intraven o u s m a g n e s i u m sulfate, I V o r s u b c u t a n e o u s b e t a m i metics, a n d o r a l p r o s t a g l a n d i n s y n t h e t a s e inhibitors. O u r e x p e r i e n c e suggests t h a t t h e e n t i r e r e g i m e n is i n a d e q u a t e for fetal surgery. C a s e 14 d e m o n s t r a t e d t h a t infusing m a g n e s i u m sulfate a n d b e t a m i m e t i c s while fluid r e s t r i c t i n g t h e m o t h e r to p r e v e n t p u l m o n a r y e d e m a severely c o m p r o m i s e s b l o o d flow to t h e p l a c e n t a a n d fetus, a n d m a y a c t u a l l y c o n t r i b u t e to p r e t e r m labor. W e l e a r n e d t h a t p r o p e r fluid m a n a g e m e n t in this c o m p l e x p o s t o p e r a t i v e s e t t i n g r e q u i r e s continuous monitoring of maternal arterial pressure a n d c e n t r a l v e n o u s p r e s s u r e , as well as f e t a l h e a r t r a t e a n d u t e r i n e c o n t r a c t i o n s . W e have n o w d e v e l o p e d t e c h n i q u e s to b e t t e r m o n i t o r t h e p r o g r e s s o f l a b o r
with radiotelemeter intrauterine pressure and uterine e l e c t r o m y o g r a m ( s u b m i t t e d for p u b l i c a t i o n ) . B u t until n o w t h e t r e a t m e n t o f p r e t e r m l a b o r h a s b e e n inadequate, primarily because the fundamental m e c h a n i s m o f l a b o r is n o t u n d e r s t o o d . T h i s final b a r r i e r m a y n o w y i e l d to t h e r e c e n t d e m o n s t r a t i o n t h a t t h e n i t r i c oxide-cyclic G M P p a t h w a y is involved in u t e r i n e r e l a x a t i o n , a n d c a n b e m a n i p u l a t e d t h e r a peutically. W i t h m a n y difficult t e c h n i c a l p r o b l e m s o v e r c o m e , t h e final b a r r i e r s to successful f e t a l r e p a i r a r e b e g i n n i n g to yield. It r e m a i n s to b e s e e n w h e t h e r t h e s e a d v a n c e s will e x t e n d t h e t i m e in u t e r o a f t e r f e t a l r e p a i r e n o u g h to i m p r o v e o u t c o m e for C D H . W e b e l i e v e t h a t t h e efficacy, safety, a n d c o s t - e f f e c t i v e n e s s o f r e p a i r b e f o r e b i r t h s h o u l d b e d i r e c t l y c o m p a r e d to o p t i m a l p o s t n a t a l c a r e i n c l u d i n g E C M O in a p r o s p e c tive trial. A f t e r extensive review, t h e N a t i o n a l Instit u t e s o f H e a l t h h a s a g r e e d to f u n d this trial.
REFERENCES 1. Harrison MR, Bjordal RI, Langmark F, et al: Congenital diaphragmatic hernia: The hidden mortality. J Pediatr Surg 13:227230, 1978 2. Adzick NS, Vacanti JP, Lillehei CW, et al: Fetal diaphragmatic hernia: Ultrasound diagnosis and clinical outcome in 38 cases. J Pediatr Surg 24:654-658, 1989 3. Benacerraf BR, Adzick NS: Fetal diaphragmatic hernia: Ultrasound diagnosis and clinical outcome in 19 cases. Am J Obstet 156:573-576, 1987 4. Adzick NS, Harrison MR, Glick PL, et al: Diaphragmatic hernia in the fetus: Prenatal diagnosis and outcome in 94 cases. J Pediatr Surg 20:357-361, 1985 5. Nakayama DK, Harrison MR, Chinn DH, et al: Prenatal diagnosis and natural history of the fetus with a congenital diaphragmatic hernia: Initial clinical experience. J Pediatr Surg 20:118-124, 1985 6. Wilson JM, Lund DP, Lillehei CW, et al: Congenital diaphragmatic hernia: Predictors of severity in the ECMO era. J Pediatr Surg 26:1028-1034, 1991 7. O'Rourke PP, Lillehei CW, Crone RK, et al: The effect of extracorporeal membrane oxygenation on the survival of neonates with high risk congenital diaphragmatic hernia: 45 cases from a sing!e institution. J Pediatr Surg 26:147-152, 1991 8. Wilson JM, Lund DP, Lillehei CW, et al: Delayed repair and preoperative ECMO does not improve survival in high-risk congenital diaphragmatic hernia. J Pediatr Surg 27:368-375, 1992 9. Price MR, Galantowicz ME, Stolar CJH: Congenital diaphragmatic hernia, extracorporeal membrane oxygenation, and death: A spectrum of etiologies. J Pediatr Surg 26:1023-1027, 1991 10. Atkinson JB, Ford EG, Humphries B, et al: The impact of extracorporeal membrane support in the treatment of congenital diaphragmatic hernia. J Pediatr Surg 26:791-793, 1991 11. Breaux CW, Rouse TM, Cain WS, et al: Improvement in survival of patients with congenital diaphragmatic hernia utilizing a strategy of delayed repair after medial and/or extracorporeal membrane oxygenation stabilization. J Pediatr Surg 26:333-338, 1991 12. Harrison MR, Jester JA, Ross NA: Correction of congenital diaphragmatic hernia in utero. I. The model: Intrathoracic balloon produces fatal pulmonary hypoplasia. Surgery 88:174-182, 1980
13. Harrison MR, Bressack MA, Churg AM, et al: Correction of congenital diaphragmatic hernia in utero. II. Simulated correction permits fetal lung growth with survival at birth. Surgery 88:260-268, 1980 14. Harrison MR, Ross NA, deLorimier AA: Correction of congenital diaphragmatic hernia in utero. III. Development of a successful surgical technique using abdominoplasty to avoid compromise of umbilical blood flow. J Pediatr Surg 16:934-942, 1981 15. Adzick NS, Outwater K, Harrison MR, et al: Correction of congenital diaphragmatic hernia in utero. IV. An early gestational fetal lamb model for pulmonary vascular morphometric analysis. J Pediatr Surg 20:673-680, 1985 16. Harrison MR, Langer JC, Adzick NS, et al: Correction of congenital diaphragmatic hernia in utero. V. Initial clinical experience. J Pediatr Surg 25:47-57, 1990 17. Vacanti JP, O'Rourke PP, LiUehei CW, et al: The cardiopulmonary consequences of high-risk diaphragmatic hernia. Pediatr Surg Int 3:1-5, 1988 18. Geggel RL, Murphy JD, Langleben D, et al: Congenital diaphragmatic hernia: Arterial structural changes and persistent pulmonary hypertension after surgical repair. J Pediatr 107:457464, t985 19. Bohn D, Tanura M, Perrin D, et al: Ventilatory predictors of pulmonary hypoplasia in congenital diaphragmatic hernia, confirmed by morphologic assessment. J Pediatr 111:423-431, 1987 20. Tibboel D, Bos AP, Pattenier JW, et al: Preoperative stabilization with delayed repair in congenital diaphragmatic hernia. Z Kinderchir 44:139-143, 1989 21. Beals DA, Schlou BL, Vacanti JP, et al: Pulmonary growth and remodeling in infants with high risk congenital diaphragmatic hernia. J Pediatr Surg 27:997-1002, 1992 22. Hatch El, Kendall J, Blumhagen J: Stomach position as an in utero predictor of neonatal outcome in left-sided diaphragmatic hernia. J Pediatr Surg 27:778-780, 1992 23. Harrison MR, Adzick NS, Longaker MT, et al: Successful repair in utero of a fetal diaphragmatic hernia after removal of viscera from the left thorax. N Engl J Med 322:1522-1524, 1990 24. Harrison MR, Anderson J, Rosen MA, et al: Fetal surgery in
DIAPHRAGMATIC HERNIA IN UTERO
1417
the primate. I. Anesthetic, surgical, and tocolytic management to maximize fetal-neonatal survival. J Pediatr Surg 17:115-122, 1982 25. Nakayama KD, Harrison MR, Seron-Ferre M, et al: Fetal surgery in the primate. II. Uterine electromyographic response to operative procedure and pharmacologic agents. J Pediatr Surg 19:333-339, 1984 26. Adzick NS, Harrison MR, Glick PL, et al: Fetal surgery in the primate. III. Maternal outcome after fetal surgery. J Pediatr Surg 21:477-480, 1986 27. 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 28. Harrison MR, Adzick NS, Flake AW, et al: The CDH two-step: A dance of necessity. J Pediatr Surg (in press) 29. Hedrick MH, Jennings RW, MacGillivray TE, et al: Endo-
scopic catheterization of placental vessels for chronic fetal vascular access. Surg Forum 43:504-505, 1992 30. Jennings RW, Adzick NS, Longaker MT, et al: New techniques in fetal surgery. J Pediatr Surg 27:1329-1333, 1992 31. Jennings RW, MacGillivray TE, Rudolph AM, et al: Vascular circulation in correction of congenital diaphragmatic hernia. Surg Forum 43:624-631, 1992 32. Bond SJ, Harrison MR, Slotnick RN, et al: Cesarean delivery and hysterotomy using an absorbable stapling device. Obstet Gyneco174:25-28, 1989 33. Glick PL, Stannard VA, Leach CL, et al: Pathophysiology of congenital diaphragmatic hernia II: The fetal lamb CDH model is surfactant deficient. J Pediatr Surg 27:382-388, 1992 34. Sabik JF, Assad RS, Hanley FL: Halothane as an anesthetic for fetal surgery. J Pediatr Surg 28:542-547, 1993
Discussion R. Arensman (Chicago, IL): Sixty-one patients were evaluated for this surgery, only 14 went to surgery. You've indicated that the remaining 47 went on to deliver and had an overall survival of 40%. Could you give us any more information about those 47 and did they receive what we would consider the most aggressive other alternative therapy such as a chance at extracorporeal membrane support? Four children survived of the 14, which is a 29% survival rate. This is still less than the survival rate for the 47 who did not come to fetal surgery. How do we factor this into the equation when recommending that families go to San Francisco to enter into the trial? Only four of the 14 children that entered the study had polyhydramnios, although all of these children (61) had an ultrasound diagnosis made before 24 or 25 weeks. We would expect this group to be in the 80% to 85% mortality group. But if this is much less, as has been suggested by some of the E C M O centers, then how do we stack this therapy up against conventional delivery and E C M O support? There were two maternal complications, I believe you addressed very nicely how we're going to try and avoid these in the future, but both of those mothers returned for a second operative procedure. How do we factor this into the equation and what do we recommend to these mothers as far as the risk to themselves? Finally, I would make a comment that the manuscript very nicely points out the tremendous financial problems we're going to have in this regard. Dr Harrison documents that only five insurance companies were willing to pay for any of this care. Although the vast majority of these patients were privately insured, only two paid in full. Thirty-three percent to
80% of the patients had some amount of the hospital cost paid through their own private resources; the medical center had to bear the rest of this burden. Certainly the N I H grant will solve some of these problems, but a comment on the cost of this modality and how it will be borne might be helpful to this group in making referrals. The problems remain formidable, but the UCSF group has slowly progressed and continues to report innovative solutions to each of the ongoing problems. M. Harrison (San Francisco, CA): The issue about polyhydramnios is interesting because there's no question that it is a risk factor. But it's not very helpful in decision making at 24 weeks because it's seldom present at 24 weeks even if it's going to be present at 28 or 30 weeks. Most of these babies don't have polyhydramnios at 24 weeks so we can't actually use it as a risk factor at that age. The other risk factors have proven reasonable. M o n e y - - a tough problem we've been struggling with for a good bit of time. We've had a lot of support from our institution, but it can't go on forever. The N I H trial is a wonderful temporary solution. It took us 3 years of reviews and revisions, but actually having it approved and funded is a tremendous vote of confidence in the enterprise and in our attempt to settle the issue. Maternal complications--a potentially troublesome issue. I think you're asking about one of the survivors where we went back about ten days after she developed oligohydramnios. We diagnosed a uterine leak, went back and closed the leak, and that fetus was in fact a survivor. That's the only one we reexplored. Fortunately, we've had no significant problems long-term with the mothers. One problem that's guaranteed is the significant morbidity of toco-
1418
lytic therapy. It's worse than the operation. And that's why it is the last barrier to fetal intervention. If it can be overcome, a new era will begin. ECMO. The babies who didn't have fetal surgery but went on had a 60% mortality. They have been analyzed in great detail elsewhere. What happened to
HARRISON ET AL
them taught us some very interesting lessons about fetal demise, about preterm delivery from polyhydramnios, and about immediate postnatal demise that prevented them from going on ECMO. These were optimally treated babies. They all had ECMO available.
O N G E N I T A L diaphragmatic hernia (CDH) remains a frustrating clinical problem. Although less severely affected babies survive with modern postnatal surgical care including extracorporeal membrane oxygenation (ECMO), the majority of babies who are severely affected die despite all intervention. 1-n Because retrospective estimates of mortality for C D H vary widely and are flawed by a "hidden mortality" of unknown magnitude, we did a prospective study which shows the mortality for fetuses with correctable isolated diaphragmatic hernias diagnosed before 24 weeks is 60% despite optimal postnatal care including E C M O (submitted for publication). Salvage of these severely affected babies with isolated diaphragmatic hernias is an unsolved problem. The pulmonary hypoplasia that limits survival is an underdevelopment of both the parenchyma and the pulmonary vascular bed, 12-19and appears to be reversible both experimentally and clinically. 13,15,21,22However, many weeks or even months are required. E C M O is limited to 1 to 2 weeks, and while useful for marginal babies, it clearly cannot salvage severely affected babies. Long-term support or replacement of lung function after birth will require either an artifi-
C
Journal of Pediatric Surgery, Vo128, No 10 (October), 1993: pp 1411-1418
cial placenta or neonatal lung transplantation. Repair before birth with continued support on the "placental E C M O " while the fetal lung grows and recovers would be ideal. However repair in utero with continued gestation has proven to be a formidable challenge.16, 23 Because maternal safety is an overriding issue for fetal surgery, we first demonstrated in the nonhuman primate 24-26 and then in our first 17 patients 16,17 that hysterotomy and fetal surgery did not adversely affect the mother or her ability to have subsequent normal children. Our initial attempts to repair six fetuses with diaphragmatic hernia showed that although the repair was feasible, there were very significant technical difficulties particularly from herniated liver. 16 We now report that further experimental and clinical studies over the last 3 years have resolved many of the intraoperative technical problems 2s,29 and allowed a few successful repairs. However, most attempts were not successful, and this frustrating and disappointing experience has taught us many hard-earned lessons about deficiencies in our intraoperative and postoperative management, particularly our inability to effectively monitor and treat preterm labor. MATERIALS AND METHODS
Sixty-one patients were referred for consideration of in utero repair of prenatally diagnosed diaphragmatic hernia between January 1989and December 1991.Fetal repair was attempted in 14 with severe isolated lesions diagnosed before 24 weeks' gestation. All had normal chromosomes by amniocentesis or percutaneous umbilical blood sampling, had left-sided hernias with dilated stomach incarcerated in the chest and significant contralateral mediastinal shift. Only 4 had polyhydramnios.Although polyhydramnios correlates with poor outcome, it frequently appears later than the 22 to 26 weeks when these fetuses were evaluated. In all cases the familywas interviewedand counseled by a perinatologist,
From the Fetal Treatment Center and the Departments of Surgery, Obstetrics, Gynecology, Reproductive Sciences, Anesthesiology, Radiology, and Nursing, University of California, San Francisco, CA. Presented at the 1992 Annual Meeting of the Section on Surgery of the American Academy of Pediatrics, San Francisco, California, October 9-11, 1992. Address reprint requests to Michael R. Harrison, MD, Fetal Treatment Center, University of California, San Francisco, 513 Parnassus Ave, Room 585-HSE, San Francisco, CA 94143-0570. Copyright 9 1993 by W.B. Saunders Company 0022-3468/93/2810-0037503.00/0 1411
1412
HARRISON ET AL
Table 1. Fetal Diaphragmatic Hernia Repair: Analysis of Cases Case No. Survived
1
4
6
12
Died at operation
2
5
8
10
11
Repaired, died in utero
3
Clinical Presentation
Procedure
Course
27 yo G2P1; 24.5 wk Posterior fundal hysterotomy; subcostal exmale; liver: lateral posure, diaphragm patch, abdominal segment up; plapatch; HR 120-150; 54 min centa: right anterior
Preterm labor ~ PROM; C/S at 32 wk; 1,920 g, Apgar 7 and 9; resp support 1 mo; fundoplication at 4 mo Posterior fundal hysterotomy; Subcostal Preterm labor ~ PROM; 27 yo G3P2; 25 wk feexposure; diaphragm patch, lined abC/S at 32 wk; 1,850 g, male; liver: down; Apgar 7 and 9; Rasp placenta: midline dominal patch; stable HR 130-150; 76 rain anterior support 2 da Posterior fundal hysterotomy; difficult fetal Pulmonary edema = toco30 yo G2P0; 24 wk feposition --~ bradycardia; diaphragm patch, male; liver: lateral lyrics; oligohydramnios --~ repair AF leak segment up; plaabdominal patch; HR >20; 61 min POD11; C/S at 28 wk centa: midline ante1,400 g. Apgar 3 and 8; rior rasp support 1 mo; fundoplication and repair at 2 mo 19 yo G3P1; 25.5 wk Transverse fundal hysterotomy; double nu- Benign postoperative male; liver: left lobe chal cord; "two-step" exposure: thoracic course; oligohydramup; placenta: low and abdominal exposure; diaphragm nios; C/S at 34 wk; anterior patch; abdominal patch; HR > 130; 66 2,460 g, Apgar 7 and 7; min resp support 3 wk; fundoplication at 1 mo Transverse fundal hysterotomy; difficult fe33 yo G3P2; 25 wk male; liver: left lobe tal position --* brady with manipulation; difficult to reduce incarcerated liver; braup; placenta: middys <80 x 10; unresponsive to resuscitaline anterior tion; 102 min 37 yo G3P1; 24 wk fe- Anterior hysterotomy; subcostal exposure male; liver: lateral difficult ~ subcapsular hematoma -* segment up; platransfusion; diaphragm patch; bradycarcenta: posterior dia with abdominal patch; cord puncture lateral and transfusion --~ spasm; unresponsive to resuscitation; 78 min 19 yo G2P0; 23 wk fe- Left paramedian hysterotomy; liver reducmale; liver: minimal tion ~ bradycardia ~ unresponsive to liver; placenta: latresuscitation; 85 rain eral right 33 yo G4P~; 25wk Anterior, low dght hysterotomy; velamenmale; liver: left lobe tous insertion of cord; large liver mass could not be reduced; liver resection with in chest; placenta: left fundal and anteargon beam --> sinus venosus bleed; resuscitation attempted via puncture of rior velamentous cord --> unresponsive; 100 min Anterior midline hysterotomy; Broviac cen32 yo G2P0; 24.5 wk male; liver: left lobe tral access; bradycardia from beginning ~ uterine contractions throughout; resusciup; placenta: posterior tation w/crystalloid, HCO3; liver ligated at "waist" and resected with no blood loss; uterine contractions/cord compromise --~ fetal death; 83 min
Comment Postnatal morbidity preterm delivery, need better tocolytics
Postnatal morbidity preterm delivery, need better tocolytics AF leak diagnosed by amnioinfusion and repaired, oligohydramnios 1' morbidity after birth
"Two-step" exposure excellent allowed liver reduction, need more time in utero to ~ morbidity
Autopsy: repair of diaphragm OK; kinked umbilical vein from liver reduction
Subcostal exposure inadequate "waist" in liver ---could have resected
Autopsy: "battledore" placenta, repair OK
Abdominal closure tight; cord puncture led to cord spasm
Autopsy: repair OK, .7 um- Reduction of liver ~ brabilical vein kinked dycardia (need to resect) Autopsy: sinus venosus distorted to left; cut during resection
Subcostal exposure inadequate, liver resection possible, need to locate sinus venosus by Doppler before dividing "waist"
Autopsy: repair OK, liver resection OK, no significant bleeding
"Two-step" exposure excellent, liver resection without problem, all foiled by uterine contractions, need to reassess anesthesia/tocolysis
Posterior fundal hysterotomy; core wrapped Fetus stable HR >120, 24 yo G1P0; 25 wk male; liver: left lobe around neck ~ difficult exposure; subgood anatomy by costal exposure difficult to retract liver; sonogram postoperaup anterior; plabrady <80 x 12; diaphragm patch, abtive; fetal demise 1st centa: right anterior dominal patch - ? tight; 54 min postop night; autopsy: repair OK, liver ischemia
Postoperative monitoring by tocodynamometer inadequate, ? late death umbilical distortion
DIAPHRAGMATIC HERNIA IN UTERO
1413
Table 1. Fetal Diaphragmatic Hernia Repair: Analysis of Cases (Cont'd) Case
No.
Clinical Presentation
Procedure
Course
Comment
9
25 yo G2P1; 28 wk fe- Anterior oblique hysterotomy; subcostal male; liver: left lobe exposure inadequate for liver; brady up; placenta: left <80 • 20 when liver reduced; diaphragm posterior patch, abdominal patch; 99 min
Fetus stable with good anatomy; 14 h postoperation developed bradycardia --> died; autopsy: repair intact, unexplained demise
13
39 yo G4P0; 25.5 wk Posterior fundal hysterotomy; "two-step" male; liver: left lobe exposure: excellent; lateral segment left up, good waist; plalobe resected without problems; diacenta: anterior midphragm patch; abdominal silo; fetal HR line stable > 140 throughout; uterine contractions problem; 69 rain
Stable fetus--good HR; Fetal liver resection feasi30 h postoperation deble, postoperative manveloped bradycardia; agement inadequate, unresponsive to atfetal monitor OK, monitempted resuscitation tor of labor poor, tocoincl PUBS; autopsy: lytics (MgSO4 + terburepair intact, no bleedtaline) toxic, need ing, Hct 43 maternal/fetal ICU AF volume remained low; Tocolysis inadequate vaginal leak and PROM contractions --> PROM, at 26 wk; C/S at 31 wk; oligohydramnios --* 1,425 g, Apgars 3 and lung hypopiasia 6; died post 2 h; autopsy: pulmonary hypoplasia, small disruption diaphragm Difficult postoperative Intraoperative problems course due to toxicity solved; postoperative of tocolytics; preterm management inadlabor poorly controlled; equate; tocolysis AF vaginal leak --, (MgSO4 + terbutaline) PROM at 28 wk; C/S at toxic; need new anes31 wk; 1,565 g, Apgars thetic/tocolytic man3 and 7; died post 2 wk agement + fetal ICU
Died after birth
29 yo G3P2; 24 wk male; liver: no liver up; placenta: posterior midline
14
Anterior transverse hysterotomy; difficult uterus tight + herniated cord; diaphragm and abdominal patch; HR stable > 120 throughout; 63 min
43 yo GzP0; 25.5 wk Low anterior, oblique hysterotomy; twomale; liver: left lobe step exposure; reduced liver with diffiup; placenta: anteculty; diaphragm patch, abdominal silo; rior left, fundal bradycardia <90 • 3 w/liver manipulation; HR stable > 120 after liver down; PUBS Hct 46, ABG OK; 79 min
Difficult but successful repair, radiotelemeter fetal ECG excellent, angiogram -~ distorted umbilical vein ? = late death
Abbreviations: G, gravida; P, para; HR, heart rate; C/S, cesarean section; Resp, respiratory; POD, postoperative day.
surgeon, counselor, anesthesiologist, clinical nurse specialist, and psychiatric social worker. Families were offered intervention only when all parties agreed that it was appropriate medically and psychologically for that family. Families signed a consent form approved by the Committee on Human Research at UCSF. Financial considerations were frequently limiting. Many families could not afford to come to San Francisco for evaluation. Although 12 of 14 had some third-party coverage, only five carriers agreed to pay for even part of the fetal procedure. Two paid in full, and the others covered 33% to 80% of hospital costs. The unreimbursed portion of the hospital bill was borne by the UCSF Medical Center and the professionals involved. More than 20 professionals donated their services through the UCSF Fetal Treatment Center. More recently, American Airlines has provided air travel though its "Miles for Kids" Program and the March of Dimes has helped the families with local accommodations. RESULTS
The clinical course and outcome of each case is presented in Table 1. The "hard-earned" lessons derived from this experience are summarized below.
Intraoperative Death (Cases 2, 5, 8, 10, and 11) Five fetuses died intraoperatively from technical problems related to reduction of incarcerated liver,
providing the following hard-earned lessons. (1) Fetal monitoring by standard electrocardiogram (ECG) leads or pulse oximeter was inadequate. The radiotelemeter ECG 3~ solved this problem. (2) The standard subcostal incision was inadequate to deal with a herniated liver. The "CDH two-step ''28 solved this problem. (3) An incarcerated liver could not be reduced without kinking the umbilical vein. Experimental work on the course of the umbilical vein 31 and on new techniques for fetal liver resection (submitted for publication) solved this problem. (4) The ductus venosus is distorted to the left by herniation and is easily injured during resection. Intraoperative Doppler to identify the course of the ductus venosus solved this problem. (5) Finally, even when all these technical maneuvers were accomplished (case 11), success was spoiled by uncontrolled uterine contractions which began as we made the hysterotomy, caused repeated bradycardias throughout the procedure, and did not respond to deepending isofluorane anesthesia. Although halogenated agents appeared to be an effective tocolytic based on our original
1414
primate work, 24-26 it was becoming clear that a new approach to both anesthesia and tocolysis was needed.
Postoperative Fetal Death (Cases 3, 9, and 13) Three had successful repairs only to die in the perioperative period. The hard-earned lessons from these frustrating cases led us to the concept of postoperative maternal-fetal intensive care. (1) Periodic fetal heart rate testing by Doppler was inadequate after fetal surgery. Bradycardia can progress quickly to fetal demise. The radiotelemeter ECG 3~ solved the postoperative as well as the intraoperative monitoring problem. (2) Monitoring of uterine activity by external tocodynamometer was inadequate to allow adjustment of tocolytics. This problem has been solved experimentally by using the radiotelemeter to monitor uterine electromyeogram and intrauterine pressure. (3) Even when fetal deterioration was accurately detected (case 13), attempted fetal resuscitation was hindered by our inability to access the fetal circulation. Recently, we have developed endoscopic catheterization of placental vessels for chronic fetal vascular access, a9 (4) Finally, postoperative tocolysis with magnesium sulfate, terbutaline, and indocin proved inadequate and even toxic.
Neonatal Death (Cases 7 and 14) After successful fetal repair, preterm labor and premature rupture of membranes led to preterm delivery and neonatal death in two cases. We learned several lessons. (1) Premature rupture of membranes and persistent oligohydramnios prevented lung growth and contributed to premature delivery and subsequent neonatal death. We then modified our technique of closing the uterus to include removal of staples, use of stays, and use of fibrin glue to help prevent amniotic fluid from dissecting under the membranes and causing a vaginal leak. (2) The standard obstetrical regimen for tocolysis was inadequate and even toxic. Case 14 had a very difficult postoperative course due to poorly controlled labor and the toxicity of magnesium sulfate and terbutaline. For the first time the patient had arterial and central venous lines placed intraoperatively and was monitored in the intensive care unit for the first 3 days. Maternal volume depletion and its adverse effects on 9uterine blood flow were documented by Doppler ultrasonography. Even after discharge from the hospital, preterm labor was poorly controlled and premature rupture of membranes was documented by a vaginal leak at 28 weeks. This case illustrated the need for improved postoperative intensive care and led us to challenge the standard obstetrical regimen for managing preterm labor with magnesium sulfate,
HARRISON ET AL
terbutaline and maternal fluid restriction. A new approach to management of the fetal-placentalmaternal unit after hysterotomy and fetal surgery would be based on reexamination of maternal-fetal anesthesia, monitoring, and tocolysis in experimental models. 29-31
Survivors (Cases 1, 4, 6, and 12) Four babies survived. Even the successful cases provided some lessons. (1) Cases 1 (reported previously23) and 4 delivered too early (32 weeks) and experienced significant postnatal morbidity which would have been minimized if pregnancy was prolonged for at least another month. (2) Amniotic fluid leak with oligohydramnios can sometimes be corrected. In case 6, the fetus did well but mother had an extremely difficult time with the tocolytics magnesium sulfate and terbutaline which led to mild pulmonary edema. Postoperative oligohydramnios was initially thought to be due to indomethacin, but an amnioinfusion showed a leak through the hysterotomy site. On postoperative day 11, the patient was reexplored and an amniotic fluid leak through a small uterine disruption was repaired. The hysterotomy wound edges looked ischemic probably secondary to the occlusive clamps used on the uterine incision at that time. This problem was solved by a special nonocclusive backbiting clamp-retractor that provides exposure and hemostasis without pressure necrosis of the uterine edges. For closure, we now remove the staples to allow approximation of the full-thickness of uterine muscle by interrupted stay sutures in addition to two layers of running Maxon, and fibrin glue to seal the inner layer. (3) Inability to control uterine contractions may have contributed to the uterine disruption and amniotic fluid leak as well as preterm labor. (4) Large diaphragmatic defects which require a prosthetic patch, whether repaired before or after birth, will often be complicated by later patch disruption with growth as well as gastroesophageal reflux. These problems once recognized can be corrected surgically as was done in cases 6 and 12. DISCUSSION
Diaphragmatic hernia is still an unsolved problem. The pathophysiologic rationale for in utero repair is compelling, 1-15 but repair of diaphragmatic hernia before birth has proven difficult. 16,23A succession of intraoperative technical problems have been slowly surmounted, only to encounter more challenges in postoperative care. The problem of accurate preoperative assessment of the extent of liver herniation has finally yielded to
DIAPHRAGMATIC HERNIA IN UTERO
careful evaluation of the umbilical vein by color Doppler. When the umbilical vein is deviated far to the left, the liver is usually up in the left chest. If the course of the sinus venosus remains below where the diaphragm should be, the herniated liver is amenable to reduction. If it appears well up in the chest, the entire liver is torqued into the chest and probably cannot be reduced without compromising umbilical flow. This is certainly true in the unusual right-sided diaphragmatic hernia as we have demonstrated in a recent case by showing that reduction of the torqued liver kinked not only the sinus venosus but also the inferior vena cava. In our experience, most fetuses with left diaphragmatic hernias have some liver herniated into the chest. This comes as a surprise, since pediatric surgeons seldom see liver in a CDH patient after birth. This is because babies who make it to surgery have already had the liver reduced from the chest by positive-pressure ventilation. Reduction of the liver after birth either by the respirator or by the surgeon causes little problems since there is no blood flow through the umbilical vein. Reduction of the liver before birth is devastating because it compromises the umbilical circulation. Herniated liver accounted for most of our failures. Disturbance of the umbilical circulation during or after liver reduction is implicated in all five intraoperative deaths and in the three early postoperative deaths. We have recently demonstrated using postmortem angiograms and vascular casts of the umbilical and hepatic circulations that kinking of the extrahepatic umbilical vessels can be prevented by dividing the falciform ligament that attaches the umbilical vein to the underside of the abdominal wall.31 In addition, we went back to the laboratory and developed techniques for fetal liver resection (submitted for publication). Now if the liver cannot be reduced without compromising umbilical circulation, the left lobe of the herniated fetal liver can be amputated as demonstrated in cases 11 and 12. The key consideration in amputating the liver is the position of the ductus venosus (see case 10) and this can now be delineated intraoperatively by Doppler ultrasound. The toughest lesson was that the subcostal incision that works so well after birth provides inadequate exposure to deal with the severe defects encountered in those fetuses. The solution was a carefully planned two-step approach using both a thoracotomy and a subcostal incision. This method allows reduction of viscera using a "push-pull" technique, reduction or amputation of incarcerated liver, and reconstruction of the diaphragm with a prosthetic patch. This technique is described in detail elsewhere. 2s
1415
Problems with uterine disruption and amniotic fluid leak (cases 6, 7, and 14) led to revision of our technique of opening and closing the gravid uterus. We have tried many ways to control bleeding from the cut uterus. Occlusive clamps made the edge of the uterus ischemic, and led to the development of a gentle back-biting, nonocclusive retractor that controis bleeding and provides exposure without causing ischemia. The stapling device we developed a decade ago prevents bleeding and seals the membranes to the myometrium. But we have now learned to remove the staples from the wound edge just prior to closure to allow muscle-to-muscle approximation. All layer stay sutures, fibrin glue, and a two-layer closure help prevent dissection of amniotic fluid resulting in vaginal leak. Even with these advances, complete repair of fetal CDH remains challenging. We have long sought another approach to improving fetal lung development that did not require the extensive manipulation necessary for total repair. The simple expedient of creating an abdominal wall defect and leaving the viscera externalized has proven unsatisfactory: decompression of the chest is inadequate and damage to the viscera can be significant, as demonstrated in one of our early cases. 16 Improving pulmonary surfactant production may be helpful for maturation33 but will not affect growth and development. Another less invasive approach appears promising. We have shown that impeding the normal egress of fetal lung fluid by controlled tracheal obstruction enlarges the hypoplastic lung and pushes the viscera back into the abdomen in fetal lambs. This new procedure, which we call "PLUG" (Plug Lung Until it Grows), may prove useful when total correction in utero is not possible. Significant obstacles to successful repair of diaphragmatic hernia and other life-threatening fetal lesions remain. The limiting factor is now anesthetic and tocolytic management. Based on our initial primate studies, 24-26 we thought that deep anesthesia with a halogenated agent (halothane, isofluorane) provided adequate anesthesia for fetus and mother as well as complete uterine relaxation. But our clinical experience has been disappointing. Indeed intraoperative contractions caused significant problems in at least half of these cases, and clearly ruined an otherwise successful repair in two (cases 1l and 13). In addition, fetal lamb studies have shown that halothane anesthesia depresses myocardial function in both the fetus and the mother and does not block the stress response that leads to placental hypoperfusion. 34We are investigating the possibility of blocking the fetal and maternal stress response with opioids. The most glaring deficiency in management is our
1416
HARRISON ET AL
inability to m a n a g e t h e p r e t e r m l a b o r i n d u c e d by h y s t e r o t o m y a n d fetal surgery. W e have b e e n f o r c e d to rely on t h e r e g i m e n u s e d ( o f t e n ineffectively) for s p o n t a n e o u s p r e t e r m labor. T h i s i n c l u d e s e x t e r n a l m o n i t o r i n g w i t h a t o c o d y n a m o m e t e r , b e d r e s t , intraven o u s m a g n e s i u m sulfate, I V o r s u b c u t a n e o u s b e t a m i metics, a n d o r a l p r o s t a g l a n d i n s y n t h e t a s e inhibitors. O u r e x p e r i e n c e suggests t h a t t h e e n t i r e r e g i m e n is i n a d e q u a t e for fetal surgery. C a s e 14 d e m o n s t r a t e d t h a t infusing m a g n e s i u m sulfate a n d b e t a m i m e t i c s while fluid r e s t r i c t i n g t h e m o t h e r to p r e v e n t p u l m o n a r y e d e m a severely c o m p r o m i s e s b l o o d flow to t h e p l a c e n t a a n d fetus, a n d m a y a c t u a l l y c o n t r i b u t e to p r e t e r m labor. W e l e a r n e d t h a t p r o p e r fluid m a n a g e m e n t in this c o m p l e x p o s t o p e r a t i v e s e t t i n g r e q u i r e s continuous monitoring of maternal arterial pressure a n d c e n t r a l v e n o u s p r e s s u r e , as well as f e t a l h e a r t r a t e a n d u t e r i n e c o n t r a c t i o n s . W e have n o w d e v e l o p e d t e c h n i q u e s to b e t t e r m o n i t o r t h e p r o g r e s s o f l a b o r
with radiotelemeter intrauterine pressure and uterine e l e c t r o m y o g r a m ( s u b m i t t e d for p u b l i c a t i o n ) . B u t until n o w t h e t r e a t m e n t o f p r e t e r m l a b o r h a s b e e n inadequate, primarily because the fundamental m e c h a n i s m o f l a b o r is n o t u n d e r s t o o d . T h i s final b a r r i e r m a y n o w y i e l d to t h e r e c e n t d e m o n s t r a t i o n t h a t t h e n i t r i c oxide-cyclic G M P p a t h w a y is involved in u t e r i n e r e l a x a t i o n , a n d c a n b e m a n i p u l a t e d t h e r a peutically. W i t h m a n y difficult t e c h n i c a l p r o b l e m s o v e r c o m e , t h e final b a r r i e r s to successful f e t a l r e p a i r a r e b e g i n n i n g to yield. It r e m a i n s to b e s e e n w h e t h e r t h e s e a d v a n c e s will e x t e n d t h e t i m e in u t e r o a f t e r f e t a l r e p a i r e n o u g h to i m p r o v e o u t c o m e for C D H . W e b e l i e v e t h a t t h e efficacy, safety, a n d c o s t - e f f e c t i v e n e s s o f r e p a i r b e f o r e b i r t h s h o u l d b e d i r e c t l y c o m p a r e d to o p t i m a l p o s t n a t a l c a r e i n c l u d i n g E C M O in a p r o s p e c tive trial. A f t e r extensive review, t h e N a t i o n a l Instit u t e s o f H e a l t h h a s a g r e e d to f u n d this trial.
REFERENCES 1. Harrison MR, Bjordal RI, Langmark F, et al: Congenital diaphragmatic hernia: The hidden mortality. J Pediatr Surg 13:227230, 1978 2. Adzick NS, Vacanti JP, Lillehei CW, et al: Fetal diaphragmatic hernia: Ultrasound diagnosis and clinical outcome in 38 cases. J Pediatr Surg 24:654-658, 1989 3. Benacerraf BR, Adzick NS: Fetal diaphragmatic hernia: Ultrasound diagnosis and clinical outcome in 19 cases. Am J Obstet 156:573-576, 1987 4. Adzick NS, Harrison MR, Glick PL, et al: Diaphragmatic hernia in the fetus: Prenatal diagnosis and outcome in 94 cases. J Pediatr Surg 20:357-361, 1985 5. Nakayama DK, Harrison MR, Chinn DH, et al: Prenatal diagnosis and natural history of the fetus with a congenital diaphragmatic hernia: Initial clinical experience. J Pediatr Surg 20:118-124, 1985 6. Wilson JM, Lund DP, Lillehei CW, et al: Congenital diaphragmatic hernia: Predictors of severity in the ECMO era. J Pediatr Surg 26:1028-1034, 1991 7. O'Rourke PP, Lillehei CW, Crone RK, et al: The effect of extracorporeal membrane oxygenation on the survival of neonates with high risk congenital diaphragmatic hernia: 45 cases from a sing!e institution. J Pediatr Surg 26:147-152, 1991 8. Wilson JM, Lund DP, Lillehei CW, et al: Delayed repair and preoperative ECMO does not improve survival in high-risk congenital diaphragmatic hernia. J Pediatr Surg 27:368-375, 1992 9. Price MR, Galantowicz ME, Stolar CJH: Congenital diaphragmatic hernia, extracorporeal membrane oxygenation, and death: A spectrum of etiologies. J Pediatr Surg 26:1023-1027, 1991 10. Atkinson JB, Ford EG, Humphries B, et al: The impact of extracorporeal membrane support in the treatment of congenital diaphragmatic hernia. J Pediatr Surg 26:791-793, 1991 11. Breaux CW, Rouse TM, Cain WS, et al: Improvement in survival of patients with congenital diaphragmatic hernia utilizing a strategy of delayed repair after medial and/or extracorporeal membrane oxygenation stabilization. J Pediatr Surg 26:333-338, 1991 12. Harrison MR, Jester JA, Ross NA: Correction of congenital diaphragmatic hernia in utero. I. The model: Intrathoracic balloon produces fatal pulmonary hypoplasia. Surgery 88:174-182, 1980
13. Harrison MR, Bressack MA, Churg AM, et al: Correction of congenital diaphragmatic hernia in utero. II. Simulated correction permits fetal lung growth with survival at birth. Surgery 88:260-268, 1980 14. Harrison MR, Ross NA, deLorimier AA: Correction of congenital diaphragmatic hernia in utero. III. Development of a successful surgical technique using abdominoplasty to avoid compromise of umbilical blood flow. J Pediatr Surg 16:934-942, 1981 15. Adzick NS, Outwater K, Harrison MR, et al: Correction of congenital diaphragmatic hernia in utero. IV. An early gestational fetal lamb model for pulmonary vascular morphometric analysis. J Pediatr Surg 20:673-680, 1985 16. Harrison MR, Langer JC, Adzick NS, et al: Correction of congenital diaphragmatic hernia in utero. V. Initial clinical experience. J Pediatr Surg 25:47-57, 1990 17. Vacanti JP, O'Rourke PP, LiUehei CW, et al: The cardiopulmonary consequences of high-risk diaphragmatic hernia. Pediatr Surg Int 3:1-5, 1988 18. Geggel RL, Murphy JD, Langleben D, et al: Congenital diaphragmatic hernia: Arterial structural changes and persistent pulmonary hypertension after surgical repair. J Pediatr 107:457464, t985 19. Bohn D, Tanura M, Perrin D, et al: Ventilatory predictors of pulmonary hypoplasia in congenital diaphragmatic hernia, confirmed by morphologic assessment. J Pediatr 111:423-431, 1987 20. Tibboel D, Bos AP, Pattenier JW, et al: Preoperative stabilization with delayed repair in congenital diaphragmatic hernia. Z Kinderchir 44:139-143, 1989 21. Beals DA, Schlou BL, Vacanti JP, et al: Pulmonary growth and remodeling in infants with high risk congenital diaphragmatic hernia. J Pediatr Surg 27:997-1002, 1992 22. Hatch El, Kendall J, Blumhagen J: Stomach position as an in utero predictor of neonatal outcome in left-sided diaphragmatic hernia. J Pediatr Surg 27:778-780, 1992 23. Harrison MR, Adzick NS, Longaker MT, et al: Successful repair in utero of a fetal diaphragmatic hernia after removal of viscera from the left thorax. N Engl J Med 322:1522-1524, 1990 24. Harrison MR, Anderson J, Rosen MA, et al: Fetal surgery in
DIAPHRAGMATIC HERNIA IN UTERO
1417
the primate. I. Anesthetic, surgical, and tocolytic management to maximize fetal-neonatal survival. J Pediatr Surg 17:115-122, 1982 25. Nakayama KD, Harrison MR, Seron-Ferre M, et al: Fetal surgery in the primate. II. Uterine electromyographic response to operative procedure and pharmacologic agents. J Pediatr Surg 19:333-339, 1984 26. Adzick NS, Harrison MR, Glick PL, et al: Fetal surgery in the primate. III. Maternal outcome after fetal surgery. J Pediatr Surg 21:477-480, 1986 27. 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 28. Harrison MR, Adzick NS, Flake AW, et al: The CDH two-step: A dance of necessity. J Pediatr Surg (in press) 29. Hedrick MH, Jennings RW, MacGillivray TE, et al: Endo-
scopic catheterization of placental vessels for chronic fetal vascular access. Surg Forum 43:504-505, 1992 30. Jennings RW, Adzick NS, Longaker MT, et al: New techniques in fetal surgery. J Pediatr Surg 27:1329-1333, 1992 31. Jennings RW, MacGillivray TE, Rudolph AM, et al: Vascular circulation in correction of congenital diaphragmatic hernia. Surg Forum 43:624-631, 1992 32. Bond SJ, Harrison MR, Slotnick RN, et al: Cesarean delivery and hysterotomy using an absorbable stapling device. Obstet Gyneco174:25-28, 1989 33. Glick PL, Stannard VA, Leach CL, et al: Pathophysiology of congenital diaphragmatic hernia II: The fetal lamb CDH model is surfactant deficient. J Pediatr Surg 27:382-388, 1992 34. Sabik JF, Assad RS, Hanley FL: Halothane as an anesthetic for fetal surgery. J Pediatr Surg 28:542-547, 1993
Discussion R. Arensman (Chicago, IL): Sixty-one patients were evaluated for this surgery, only 14 went to surgery. You've indicated that the remaining 47 went on to deliver and had an overall survival of 40%. Could you give us any more information about those 47 and did they receive what we would consider the most aggressive other alternative therapy such as a chance at extracorporeal membrane support? Four children survived of the 14, which is a 29% survival rate. This is still less than the survival rate for the 47 who did not come to fetal surgery. How do we factor this into the equation when recommending that families go to San Francisco to enter into the trial? Only four of the 14 children that entered the study had polyhydramnios, although all of these children (61) had an ultrasound diagnosis made before 24 or 25 weeks. We would expect this group to be in the 80% to 85% mortality group. But if this is much less, as has been suggested by some of the E C M O centers, then how do we stack this therapy up against conventional delivery and E C M O support? There were two maternal complications, I believe you addressed very nicely how we're going to try and avoid these in the future, but both of those mothers returned for a second operative procedure. How do we factor this into the equation and what do we recommend to these mothers as far as the risk to themselves? Finally, I would make a comment that the manuscript very nicely points out the tremendous financial problems we're going to have in this regard. Dr Harrison documents that only five insurance companies were willing to pay for any of this care. Although the vast majority of these patients were privately insured, only two paid in full. Thirty-three percent to
80% of the patients had some amount of the hospital cost paid through their own private resources; the medical center had to bear the rest of this burden. Certainly the N I H grant will solve some of these problems, but a comment on the cost of this modality and how it will be borne might be helpful to this group in making referrals. The problems remain formidable, but the UCSF group has slowly progressed and continues to report innovative solutions to each of the ongoing problems. M. Harrison (San Francisco, CA): The issue about polyhydramnios is interesting because there's no question that it is a risk factor. But it's not very helpful in decision making at 24 weeks because it's seldom present at 24 weeks even if it's going to be present at 28 or 30 weeks. Most of these babies don't have polyhydramnios at 24 weeks so we can't actually use it as a risk factor at that age. The other risk factors have proven reasonable. M o n e y - - a tough problem we've been struggling with for a good bit of time. We've had a lot of support from our institution, but it can't go on forever. The N I H trial is a wonderful temporary solution. It took us 3 years of reviews and revisions, but actually having it approved and funded is a tremendous vote of confidence in the enterprise and in our attempt to settle the issue. Maternal complications--a potentially troublesome issue. I think you're asking about one of the survivors where we went back about ten days after she developed oligohydramnios. We diagnosed a uterine leak, went back and closed the leak, and that fetus was in fact a survivor. That's the only one we reexplored. Fortunately, we've had no significant problems long-term with the mothers. One problem that's guaranteed is the significant morbidity of toco-
1418
lytic therapy. It's worse than the operation. And that's why it is the last barrier to fetal intervention. If it can be overcome, a new era will begin. ECMO. The babies who didn't have fetal surgery but went on had a 60% mortality. They have been analyzed in great detail elsewhere. What happened to
HARRISON ET AL
them taught us some very interesting lessons about fetal demise, about preterm delivery from polyhydramnios, and about immediate postnatal demise that prevented them from going on ECMO. These were optimally treated babies. They all had ECMO available.