The diagnosis of abnormalities before birth has rapidly expanded during the past 15 years, primarily because of the widespread use of prenatal ultrasonography. Reports of in utero ultrasonographic diagnosis of congenital anomalies in the 1970s led to the accurate diagnosis before birth of a number of abnormalities potentially correctable by surgery. As a result, the fetus can be regarded as a patient.

Although correctable malformations diagnosed prenatally are best managed by appropriate medical and surgical therapy after planned delivery at term, life-threatening anatomical problems (e.g. diaphragmatic hernia) may be amenable to direct surgical repair before birth. Open fetal surgery was accomplished after the pathophysiology of the disorder had been defined in animal models; the natural history of the disease had been established by following untreated cases; selection criteria for intervention were developed; and the safety of the proposed procedures for both mother and fetus was demonstrated in a non-human primate model. This last condition was the most important, since the principal deterrent to direct intervention by hysterotomy is risk to the mother.

Unrelieved urinary tract obstruction interferes with fetal development, and the resulting oligohydramnios secondary to decreased fetal urine output produces pulmonary hypoplasia, which is often fatal at birth.. Life-threatening problems of respiratory and renal insufficiency may be ameliorated if the obstruction is relieved early enough in gestation to allow normal development to proceed.

Obstruction of the urinary tract early in development of the fetal lamb leads to renal dysplasia, mimicking the disease in human neonates; this can be reversed by correction later in gestation. The oligohydramnios-induced pulmonary hypoplasia associated with obstructive uropathy in lambs is similar to that seen in human fetuses, and decompression of the obstructed urinary tract has been shown to restore amniotic fluid volume and allow compensatory lung growth.

Experience with 20 human fetuses having congenital bilateral hydronephrosis has generated four criteria that classify the fetus as having good or poor prognosis (Table 1) 577. The criteria for a good prognosis (urine Na <100 mEq/l, Cl <90 mEq/l, osmolality <210 mosmol, and normal fetal kidneys by ultrasound) reliably predict neonatal and long-term outcome after in utero urinary tract decompression. Urine for laboratory studies can be obtained by a single bladder aspiration.

Ultrasound should be used to confirm the diagnosis, the anatomical level of obstruction, the status of the amniotic fluid, and the presence of associated anomalies. If an associated life-threatening anomaly is present, the family should be counselled appropriately and allowed to choose expectant management or termination of the pregnancy. If hydronephrosis appears to be an isolated defect and the amniotic fluid volume is adequate, serial ultrasound examinations should be obtained. If amniotic fluid volume remains adequate, then the mother should receive routine obstetrical care, and the fetus can be treated postnatally.

If moderate to severe oligohydramnos develops, a complete prognostic evaluation should be undertaken to determine the potential of the fetus for normal renal and pulmonary function at birth. Aggressive obstetrical care or in utero decompression is not indicated for the fetus with predicted renal dysplasia. There are two management options for the fetus likely to have good renal function depending on fetal lung maturity. If the lungs are mature, immediate delivery and ex utero decompression is indicated; immature lungs are an indication for in utero decompression. For open fetal surgery, the gestational age at diagnosis should be less than 28 weeks, as successful decompression can occasionally be performed after this time with a percutaneously placed fetal bladder catheter. Prior to 28 weeks' gestation, fetal catheter placement has been unsuccessful because of frequent obstruction and dislodgement.

Only 7 of more than 200 cases of bilateral hydronephrosis referred for management over the last decade have been deemed appropriate for open surgery. The first fetus was treated by bilateral ureterostomies; the subsequent six underwent marsupialization of the bladder at 18 to 24 weeks' gestation. The first five pregnancies proceeded to caesarean delivery at 32 to 35 weeks' gestation. The sixth fetus never drained urine well after ureterostomy and was removed at re-exploration at the parent's request because of a severe cloacal anomaly. The seventh fetus underwent successful urinary tract decompression, but died 2 weeks later from pulmonary hypoplasia. This baby was delivered prematurely when the mother discontinued her oral tocolytic therapy.

Normal amniotic fluid dynamics developed in three pregnancies and all three babies had adequate pulmonary function at birth, suggesting that the pulmonary hypoplasia associated with early severe oligohydramnios had been reversed. Two others died at birth with pulmonary hypoplasia. No amniotic fluid developed in one pregnancy, even after decompression, reflecting our initial inability to predict fetal renal function. The other neonate had some amniotic fluid after decompression, but a tiny chest cavity due to a long period of severe oligohydramnios before decompression. Of the three surviving infants, one had normal renal function when she died of unrelated causes at 9 months of age. One has normal renal function at age 3 years. Renal function began to fail in the third child by 3 years; this child developed normally, and subsequently received a successful kidney transplant. Our current selection criteria would now accurately exclude from treatment the two fetuses who died of pulmonary hypoplasia.

Improvements in selection now make it possible to avoid intervention in hopeless cases. This small series of human cases confirms that the development of fatal pulmonary hypoplasia can be prevented if amniotic fluid dynamics can be restored by decompression of the obstructed urinary tract. It is still unclear whether in utero intervention arrested or reversed cystic dysplasia changes caused by obstructive uropathy, since it is possible that the dysplastic changes initiated in utero will compromise renal function progressively as functional demand increases with growth. We believe, however, that relief of obstruction during the most active phase of nephrogenesis, between 20 and 30 weeks' gestation, may prevent further damage and allow nephrogenesis to proceed normally. Further experience and long-term follow-up are necessary to determine the effectiveness of in utero decompression in reversing or arresting renal damage caused by obstruction.

Congenital diaphragmatic hernia is an anatomically simple defect easily correctable by removing the herniated viscera from the chest and closing the diaphragm. However, many infants with congenital diaphragmatic hernia die of pulmonary insufficiency despite optimal postnatal care because their lungs are too hypoplastic to support extrauterine life. The pulmonary hypoplasia associated with congenital diaphragmatic hernia has been well documented clinically and experimentally; it appears to be caused by compression of the developing fetal lung by herniated bowel.

Management of the fetus with prenatally diagnosed congenital diaphragmatic hernia requires an understanding of the natural history, pathophysiology, and prognostic factors which affect outcome. Our multicentre survey demonstrated that prenatal diagnosis of congenital diaphragmatic hernia is accurate, the mortality rate is high (80 per cent), and polyhydramnios is a prenatal predictor of poor clinical outcome. Even with the use of extracorporeal membrane oxygenation to maximize postnatal support, survival of affected babies is poor.

Extensive experience has demonstrated the efficacy, feasibility, and safety of repair in utero in both lambs and monkeys. We therefore attempted to treat seven highly selected fetuses with severe congenital diaphragmatic hernia by open surgery (Fig. 2) 2218. The first three fetuses died during surgery following unsuccessful attempts to reduce the friable incarcerated liver from the fetal chest. In the fourth case, a Gore-Tex diaphragm was constructed around the liver, but lung decompression was ineffective and the baby died at birth. Surgery in the last three fetuses was successful, and the technical problem with the herniated liver proved to be surmountable. All three demonstrated rapid growth of the lung in utero, and had surprisingly good lung function after birth. Two subsequently died of non-pulmonary problems (an unrelated nursery accident in one and intestinal complications in the other), but the last baby has done well.

Figure 3 2219 shows our proposed algorithm for management. Once a diaphragmatic hernia is diagnosed by ultrasound, amniocentesis (results in about 2 weeks) or percutaneous umbilical blood sampling (results in about 2 days) should be performed to allow chromosomal analysis. Screening for other anatomical abnormalities should be undertaken by an experienced obstetrical sonographer, and cardiac abnormalities should be sought, using fetal echocardiography. When associated serious anomalies are discovered, the family may choose to terminate the pregnancy.

There is a wide spectrum of severity in congenital diaphragmatic hernia. Some mildly affected fetuses detected later in gestation develop polyhydramnios late or not at all, and have a small volume of viscera in the chest. These fetuses should be followed by sonogram and delivered at an appropriate tertiary perinatal centre after the lungs are mature. Unfortunately, the majority of fetuses have severe disease and do not survive even with optimal conventional pre- and postnatal management. In general, these fetuses are detected early, develop polyhydramnios early, and have a larger volume of viscera in the chest (dilated stomach, impressive mediastinal shift, little lung visible in either thorax). Every effort should be made to determine whether the liver is herniated into the chest—this affects prognosis and the technical difficulty of repair before birth. Between 20 and 30 weeks of gestation the family can choose between conventional management or fetal repair depending on the results of assessment and their personal attitude. After 30 weeks, the only option is conventional management, aimed at maximizing postnatal care. Although fetal repair is theoretically possible until 32 weeks, it is probably best performed before 30 weeks. First, the longer the decompressed lung has to grow before it is required to support life at birth, the better (and preterm labour associated with hysterotomy may shorten the remaining gestation). Second, it is our impression that the risk of inducing preterm labour is greater when the procedure is performed later in gestation. Although it would seem technically advantageous to operate on a larger fetus with more mature tissues, surgical procedures are feasible in fetuses as young as 18 weeks. At present, fetuses between 20 and 30 weeks' gestation are considered suitable candidates for in utero surgery.

Wound healing in the fetus is fundamentally different from that process in the adult: the fetus has the remarkable ability to heal without scarring or fibrosis. The mechanisms that underlie these qualities may be unique to the fetal cells, the fetal environment, or a combination of the two.

One of the major differences between fetal and adult wounds lies in the wound extracellular matrix. The fetal wound extracellular matrix is rich in hyaluronic acid, a glycosaminoglycan found in high concentrations whenever there is rapid cell proliferation and movement. A hyaluronic acid-stimulating activity is present in fetal wound fluid, fetal serum, and amniotic fluid, but not in adult wound fluid. This substance may account for the presence of hyaluronic acid in the extracellular matrix of fetal wounds; this in turn may provide an environment capable of supporting regenerative-like healing without scar formation.

More work is needed to determine the role of the fetal environment (i.e. low arterial po&sub2;, amniotic fluid exposure), growth factor expression of fetal cells, and the role of hyaluronic acid, alone and in complexes with protein, in organizing the fetal wound matrix. Fetal surgeons may in the future be able to take advantage of scarless fetal healing and repair certain craniofacial abnormalities such as cleft lip in utero. The real benefit of this rapidly expanding area of research, however, is in understanding the unique scarless fetal healing process so that wound healing in children and adults can be engineered to be more fetal-like.

Adzick NS, et al. Diaphragmatic hernia in the fetus: prenatal diagnosis and outcome in 94 cases. J Pediatr Surg 1985; 20: 357–61.

Adzick NS, et al. Correction of congenital diaphragmatic hernia in utero IV. An early gestational fetal lamb model for pulmonary vascular morphometric analysis. J Pediatr Surg 1985; 20: 673–80.

Adzick NS, et al. Fetal surgery in the primate. III. Maternal outcome after fetal surgery. J Pediatr Surg 1986; 21: 477–80.

Crombleholme TM, et al. Early experience with open fetal surgery for congenital hydronephrosis. J Pediatr Surg 1988; 23: 1114–21.

Glick PL, Harrison MR, Adzick NS, Noall RA, Villa RL. Correction of congenital hydronephrosis in utero IV. In utero decompression prevents renal dysplasia. J Pediatr Surg 1984; 19: 649–57.

Glick PL, et al. Management of the fetus with congenital hydronephrosis II: Prognostic criteria and selection for treatment. J Pediatr Surg 1985; 20: 376–87.

Harrison MR, Ross NA, deLorimier AA. Correction of congenital diaphragmatic hernia in utero III. Development of a successful technique using abdominoplasty to avoid compromise of umbilical blood flow. J Pediatr Surg 1981; 16: 934–42.

Harrison MR, et al. Fetal surgery for hydronephrosis. N Engl J Med 1982; 306: 591–3.

Harrison MR, et al. Fetal surgery in the primate. I. Anesthetic, surgical and tocolytic management to maximize fetal-neonatal survival. J Pediatr Surg 1982; 7: 115–22.

Harrison MR, Filly RA, Golbus MS, eds. The Unborn Patient. Orlando: Grune & Stratton, Inc., 1984.

Harrison MR, et al. The correction of congenital diaphragmatic hernia in utero. V. Initial clinical experience. J Pediatr Surg 1990; 25: 47–57.

Harrison MR, et al. Successful repair in utero of a fetal diaphragmatic hernia after removal of herniated viscera from the left thorax. N Engl J Med 1990; 322: 1582–4.

Longaker MT, et al. Studies in fetal wound healing: IV. Hyaluronic acid-stimulating activity distinguishes fetal wound fluid from adult wound fluid. Ann Surg 1989; 667–72.

Nakayama DK, et al. Experimental pulmonary hypoplasia due to oligohydramnios and its reversal by relieving thoracic compression. J Pediatr Surg 1983; 18: 347–53.