- Email: [email protected]
Fetal Hydronephrosis: Considerations Regarding Urological Intervention
THE JOUR:NP.L Of UROLDGY
Copyright© 1984 by The 1/iliiliams & ~Nilkir.;; Co.
HYDRONEPHROSIS: CONSIDERATIOl\JS REGARDING UROLOGICAL INTERVENTION DA vm A. DIArVIOND, ROGER SANDERS AND ROBERT D. JEFFS From the James Buchanan Brady Urological Institute, and Departments of Urology and Radiology, The Johns Hopkins Hospital, Baltimore, Maryland
ABSTRACT
Real-time ultrasonic imaging during pregnancy has made it possible to diagnose accurately fetal hydronephrosis as early as at 17 weeks of gestation. The ultrasonic finding of oligohydramnios in conjunction with megacystis and biiateral hydronephrosis is the most reliable indicator of marked permanent urinary tract obstruction and, therefore, is an absolute requirement for fetal urological intervention. Fetal intervention by means of percutaneous catheter drainage of the bladder into the amniotic space is undertaken to preserve renal and pulmonary function. It appears that urinary tract decompression at 17 weeks of gestation is unable to prevent renal dysplasia. The degree to which renal cortical function is preserved by fetal drainage of the hydronephrotic kidney is unknown" Experimental evidence has suggested that early urinary tract decompression promotes improved pulmonary development, which may allow for normal pulmonary function postnatally. We report 6 cases that illustrate important considerations with regard to urological intervention for fetal hydronephrosis. Real-time ultrasonic imaging during pregnancy has made it to diagnose accurately fetal lesions of the central nervous system (obstructive hydrocephalus and myelomeningocele),1 gastrointestinal system (omphalocele and diaphragmatic hernia)2 and urinary tract (hydronephrosis, hydroureter, megacystis, multicystic kidney and polycystic kidney), 3• 4 as well as detecting oligohydramnios, As noted by Kramer,5 of primary concern to the urologist is the diagnosis of bilateral hydronephrosis in utero, which has been made as early as at l 7 to 22 weeks of gestation. a, 7 In an effort to preserve renal function prenatal drainage has been achieved by hysterotomy and percutaneously, As the number of treated patients with fetal hydronephrosis has grown the advantages, limitations and disadvantages of this form of intervention have become somewhat more clear, As Harrison and associates have stated, "The most difficult problem in prenatal management is how to select only the fetuses that might benefit from treatment . , . , Careful serial assessment of fetal anatomy and organ function is necessary to exclude those so severely affected that they cannot be saved and those so mildly affected that they will do well with standard treatment," 8 We report 6 Cases in which fetal urlT'/"HlPT)hI'(';Rl CH
intervention is ultin1ately unnecessary and in which postnatal treatment produces the desired result, Case 2. B. B. VV, underwent maternal ultrasound at 34 weeks of gestation" The study demonstrated bilateral hydronephrosis, a dilated bladder and oligohydramnios (fig, 2). The fetus proceeded to term and died at birth. Autopsy findings were consistent with the prune belly syndrome, He had a markedly stenotic urethra with significant dilatation of the bladder and distal ureters. The right kidney was multicystic and dysplastic, and the left was rudimentary with a cystic dysplastic cortex (fig, 3, A), Severe pulmonary hypoplasia also was found (fig, 3, B). Comment: This case represents the most severe form of fetal hydronephrosis in which early intervention would likely have been futile. Case 30 Maternal ultrasound late in the second trimester demonstrated bilateral hydronephrosis (fig. Subsequently, monthly ultrasound examinations revealed spontaneous resoUltrasound and an IVP vvere lution of the hv,ctr<)WcPi'.11' normal after bil"tho Comment: This case demonstrates that n-n,,,r,nrn~, may be WU~UWL~ufrom underwent maternal ultrasound at 28 demonstrated bilateral Serial studies the presence of upper tract without oligohydramnios, After delivery cesarean section, abdominal ultrasound con firmed the presence of bilateral hydronephrosis and hydroureter but demonstrated a normal-appearing bladder. A voiding cystourethrogram when the newborn was 2 days old showed a normal-appearing bladder and posterior urethra with no evidence of reflux. An IVP when the neonate was 5 days old demonstrated bilateral hydronephrosis and hydrometer. Bilateral Whitaker tests showed no evidence of functional obstruction. Cystoscopy revealed a normal-appearing urethra and bladder. Comment: This case appears to demonstrate the presence of persistent, lower tract obstruction in utero, which was not present postnatally, Management was conservative, with continued surveillance of renal function and upper tract drainage, ~,,.-u<-U>~H
Case 1. M. S, had normal maternal ultrasound in the second trimester, A followup ultrasound at 36 weeks of gestation demonstrated bilateral hydronephrosis that was greater on the right side. An excretory urogram 2 days after birth demonstrated marked right hydronephrosis with mild left ureteropelvic junction obstruction. A diethylenetriaminepentaacetic acid renal scan demonstrated 3 per cent function of the right kidney, Exploration when the newborn was 7 days old revealed a multicystic kidney with an atretic ureteropelvic junction (fig. 1, A). Pathological analysis demonstrated renal dysplasia (fig. 1, B). Comment: This case is representative of those in which fetal Accepted for publication February 7, 1984.
1155
1156
DIAMOND, SANDERS AND JEFFS
FIG. 1. Case 1. A, multicystic dysplastic kidney removed when patient was 1 week old. Multiple cysts make ultrasound differentiation from ureteropelvic junction obstruction difficult. B, histologic specimen from part A shows renal dysplasia with primitive tubules and islands of loosely woven mesenchymal tissue.
FIG. 2. Case 2. Maternal ultrasound at 34 weeks of gestation demonstrates dilatation of fetal posterior urethra (PU), bladder (B) and ureter ( U).
Case 5. B. B. S. underwent maternal ultrasound at 32 weeks of gestation. The study demonstrated oligohydramnios, a dilated bladder and bilateral hydronephrosis. At 34 weeks of gestation a double-pigtailed catheter was placed percutaneously in the fetal bladder, providing urinary drainage into the amniotic cavity. The findings at birth were consistent with severe prune belly syndrome (fig. 6). Bilateral ureterostomy was performed in the neonatal period. When the infant was 4 months old renal function was normal but pulmonary function was markedly compromised secondary to reduced tidal volume, which is consistent with the prune belly syndrome. Case 6. B. B. B. underwent maternal ultrasound at 18 weeks of gestation. The study demonstrated a dilated posterior urethra, megacystis, minimal right hydronephrosis and normal amniotic fluid volume. Percutaneous aspiration of the fetal bladder was performed and subsequent bladder filling, suggesting a normal rate of urine production, was noted on ultrasound.
FIG. 3. Case 2. A, histologic specimen of right kidney shows rudimentary glomeruli and tubules in cystic dysplastic cortex. B, histologic specimen of lung reveals rudimentary alveolar pattern with leukocytic infiltrate consistent with pulmonary hypoplasia.
Followup examination at 20 weeks of gestation demonstrated the development of left hydronephrosis in addition to the previous findings. The fetal bladder was again aspirated. At 28 weeks of gestation oligohydramnios was noted on maternal ultrasound, yet at 30 weeks the fetal bladder was of normal size with no evidence of hydronephrosis on ultrasound. The newborn was delivered at 36 weeks of gestation and required
EE7 DISCUSSION
he was 16 hours demurethral valves with marked dilatation of the ~r,~,-,,~,,,w urethra, bladder and ureters, complete bilateral cystic dysplasia of the 7, A) and hyaline membrane 7, disease of the lungs Comment: In this case an indwelling percutaneous catheter was not placed because of the development of urinary ascites at the time of initial bladder aspiration, as well as unfavorable placental positioning. It is of great interest that a dilated bladder and bilateral hydronephrosis were noted on ultrasound 10 weeks before the development of oligohydramnios. It also is noteworthy that the bladder appeared to fill at a rate indicative of normal urine production at 18 weeks of gestation and yet autopsy demonstrated complete cystic dysplasia of b~th kidneys. The ultrasound findings of normal-sized bladder and kidneys at 30 weeks of gestation after long-term dilatation suggested marked deterioration in renal function at that
FIG. 4. Case 3. Maternal ultrasound during second trimester demonstrates bilateral hydronephrosis (H) that resolved spontaneously.
The value of surgery for hvdroneDhros1s tests on our to demonstrate greater tion perhaps uuu,,v,w," intervention in utero rather than in the neonatal period. Given the frequency with which renal dysplasia is found in conjunction with obstructed renal units, 9 the preservation of renal function prenatally depends upon the prevention of renal dysplasia and cortical damage due to hydronephrosis in utero. The timing of potential fetal intervention during renal embryogenesis is critical. In the human the ureteral bud appears at 5 weeks of gestation. At 9 weeks Chwalle's membrane opens and fetal urine is first detected. At 17 weeks of gestation the
FIG. 6. Case 5. Neonate with absent abdominal musculature consistent with prune be\ly ~yndro~1e. 9oubltpi~ai!ed catheter placed percutaneously to dram ,eta! unne irom o!adaer to ammotic flmd compartment is noted in lower midline.
FIG. 5. Case 4. Maternal ultrasound (transverse section) at 36 weeks of gestation. A, hydronephrosis of kidneys. B, dilated bladder and distal ureter ( U).
1158
DIAMOND, SANDERS AND JEFFS
FIG. 7. Case 6. A, histologic specimen of cystic dysplastic kidney shows rudimentary tubules surrounded by abundance of mesenchymal tissue. B, histologic specimen of pulmonary parenchyma reveals findings consistent with hyaline membrane disease.
diagnosis of hydronephrosis is possible by ultrasound. The last division of the ureteral bud is completed at 20 weeks. Between 26 and 40 weeks of gestation the induction of nephrons occurs, and 80 to 90 per cent of the renal mass develops. At term there is a sudden increase in fetal renal blood flow secondary to a decrease in vascular resistance. 10 In 1971 Bernstein noted that urinary obstruction at the onset of differentiation of the mesonephric blastema is associated with renal dysplasia. 9 He found that unilateral ureteral anomalies are associated with ipsilateral dysplasia, lower tract obstruction is associated with bilateral dysplasia and the degree of dysplasia corresponds roughly to the severity of obstruction. In 1971 Beck demonstrated that the effect of obstruction on subsequent renal development is dependent upon the gestational age of the fetus at the onset of obstruction.11 He noted that ureteral obstruction induced at 50 to 70 days of gestation in the fetal lamb (147 days of total gestation) produced renal dysplasia, whereas obstruction during the second half of gestation produced hydronephrosis but not dysplasia (fig. 8). Roughly extrapolating from Beck's work, prevention of renal dysplasia in the human would require prenatal drainage before 14 to 20 weeks of gestation. It is only towards the end of this period that hydronephrosis can be diagnosed by ultrasound. If the present diagnostic techniques do not allow intervention at a time when renal dysplasia might be prevented, one wonders if the prenatal treatment of hydronephrosis might significantly preserve renal function. Because 80 to 90 per cent of the renal mass develops during the third trimester, one might anticipate impairment of nephron formation due to back pressure from urinary obstruction during this period. While longstanding fetal hydronephrosis undoubtedly causes thinning of the renal parenchyma, the length of time that the fetal kidney can be obstructed before irreversible damage occurs has not been demonstrated experimentally. Whether the drained fetal hydronephrotic system will undergo post-obstructive diuresis is unknown, as are the cardiovascular consequences of such a response to drainage. Pulmonary hypoplasia often precludes survival in the fetus with severe obstructive uropathy. It is associated commonly with oligohydramnios, which can be diagnosed by fetal ultrasound. The exact relationship between oligohydramnios and pulmonary hypoplasia remains controversial. It is known that amniotic fluid volume and composition are determined primarily by urine production, although pulmonary secretions and secretions of the amnion itself contribute to the total amniotic
I.O
O.S
FIRST TRIMESTER
SECOND TRIMESTER
OBSTRUCTION RESULTS IN
~
THIRD TRIMESTER
OBSTRUCTION RESULTS IN HYDRONEPHROSIS
0.6 URINE PRODUCTION 0.4
0.2
20
40
60
80
100
120
140
GESTATIONAL AGE (DAYS)
FIG. 8. Variation in urine production (ml. per minute) and pathological results of urinary tract obstruction with gestational age in fetal lamb model. 10•11
fluid. Using Sprague-Dawley rats, Symchych and Winchester demonstrated that experimental deficiency of amniotic fluid in the third trimester of pregnancy was associated with pulmonary hypoplasia. 12 Harrison and associates produced experimentally urethral obstruction in fetal lambs at 100 days of gestation with resultant hydronephrosis and pulmonary hypoplasia. 13 These authors suggest that oligohydramnios results in a small intrauterine cavity that restricts the fetal thorax and prevents expansion of the fetal lung during inspiration, resulting in pulmonary hypoplasia. In half of the lambs the urinary tract was decompressed at approximately 120 days of gestation. These animals were noted to have larger lungs that were adequate to support postnatal life, in contrast to the small lungs resulting from continued obstruction until term. Whereas the developmental insult resulting in pulmonary hypoplasia in association with congenital diaphragmatic hernia and congenital hydronephrosis appears to occur early in gestation, the experiments of Harrison and associates demonstrate that the decompressed lung late in gestation is capable of resuming normal growth and development during the last trimester when total lung growth is greatest. They advocate percutaneous drainage of urine into the amniotic fluid prenatally in the selected patient to facilitate normal pulmonary development.14, 15 Primary pulmonary hypoplasia without obvious renal, thoracic or cardiac anomalies has been observed but the volume of
1159
FETAL HYDRONEPHROSIS
amniotic fluid at delivery was not described, suggesting the possibility of an amniotic fluid leak. 16 The ultrasound finding of oligohydramnios in association with obstructive uropathy appears to be of further importance. Sanders and Graham have noted 2 cases of moderately severe hydronephrosis on serial fetal ultrasound, which resolved spontaneously, as in our case 3, with normal radiographic studies at birth. 7 In neither case was oligohydramnios present. The cause of such physiological hydronephrosis is unclear. In the fetal lamb model urine production is detected first at the beginning of the second trimester, when urinary obstruction appears to produce renal dysplasia. Fetal urine production increases to a peak early in the third trimester, subsequently decreasing before birth (fig. 8). 10 Should such a pattern of urine production appear in a human, it might account for the transient hydronephrosis noted by Sanders and Graham. Therefore, we suggest that fetal drainage be contemplated only in those cases in which oligohydramnios is present, thus, making it unlikely that the hydronephrosis is transient in nature. When a fetus has bilateral hydronephrosis and oligohydramnios one would like to assess the degree of renal function remaining before fetal drainage, with its attendant risks to the mother and child. To date, there are no good means of assessing renal function in utero. Investigators have examined bladder filling and emptying using ultrasound and have observed, with ultrasound, the response in fetal urine production to furosemide administered to the mother. 17 These methods of assessing fetal renal function fail in 2 respects. Urine production is not indicative of glomerular filtration rate and can be quite high in the adult with severe cortical loss. Indeed, in our case 6 the kidneys that underwent complete cystic dysplasia appeared to have a normal rate of urine production at 18 weeks of gestation. In addition, many of the newborns with posterior urethral valves and the prune belly syndrome have massive reflux that would distort any in utero measurement of urine production based on bladder filling and emptying. Aspiration of fetal urine is technically easy but it has not yet proved to be valuable in assessing renal function in utero. One would hope that advances in this area would provide us with a more precise means of assessing renal response to in utero drainage. The assessment of potential gain by in utero drainage becomes more critical as complications of the procedure accumulate. Those reported to date include the induction of premature labor, fetal bowel perforation, and hemorrhage and infection in the fetus and maternal uterus. Harrison and associates have presented a rational schema for managing the fetus with hydronephrosis. 8 Apparent unilateral hydronephrosis (secondary to ureteropelvic junction obstruction, multicystic kidney and primary megaureter) and bilateral hydronephrosis without oligohydramnios do not require in utero treatment. We hasten to add that in cases of unilateral hydronephrosis the presence of a contralateral, normal renal unit should be established. If bilateral hydronephrosis and megacystis are accompanied by marked oligohydramnios, decompression of the urinary tract and restoration of normal amniotic fluid dynamics should be undertaken. If the fetus is too immature for delivery and postnatal therapy, percutaneous catheter decompression of the bladder into the amniotic fluid should be performed. The presence of oligohydramnios is an absolute requirement for intervention, since it is the most reliable indicator of marked permanent obstruction. However, the more severe the oligohydramnios, the worse the prognosis for recovery of renal and pulmonary function. In cases in which oligohydramnios appears questionable in nature serial ultrasound should be performed
to assess amniotic fluid volume. Immediately upon birth these newborns should be evaluated thoroughly urologically to undergo appropriate corrective procedures. At present, prenatal intervention seems most justified when marked bilateral hydronephrosis presents with oligohydramnios. Under these circumstances percutaneous drainage of the urinary tract restores amniotic fluid dynamics and attempts to preserve renal and pulmonary function. The success of fetal intervention for hydronephrosis relies upon our improved un derstanding of fetal renal physiology as well as the pathogenesis of renal dysplasia, fetal hydronephrosis and pulmonary hypoplasia. In addition, we must demonstrate experimentally that in utero drainage is of significant benefit to embryogenesis of the kidneys and lungs. REFERENCES 1. Hidalgo, H., Bowie, J., Rosenberg, E. R., Ram, P. C., Ford, K. and Lipsit, E.: In utero sonographic diagnosis of fetal cerebral anomalies. Amer. J. Rad., 139: 143, 1982. 2. Jassani, M. N., Gauderer, M. W. L., Fanaroff, A. A., Fletcher, B. and Merkatz, I. R.: A perinatal approach to the diagnosis and management of gastrointestinal malformations. Obst. Gynec., 59: 33, 1982. 3. Hadlock, F. P., Deter, R. L., Carpenter, R., Gonzalez, E. T. and Park, S. K.: Sonography of fetal urinary tract anomalies. Amer. J. Rad., 137: 261, 1981. 4. Chinn, D. H. and Filly, R. A.: Ultrasound diagnosis of fetal genitourinary tract anomalies. Urol. Rad., 4: 115, 1982. 5. Kramer, S. A.: Current status of fetal intervention for congenital hydronephrosis. J. Urol., 130: 641, 1983. 6. Garrett, W. J., Kossoff, G. and Osborne, R. A.: The diagnosis of fetal hydronephrosis, megaureter and urethral obstruction by ultrasonic echography. Brit. J. Obst. Gynaec., 82: 115, 1975. 7. Sanders, R. and Graham, D.: Twelve cases of hydronephrosis in utero diagnosed by ultrasonography. J. Ultrasound Med., 1: 341, 1982. 8. Harrison, M. R., Filly, R. A., Golbus, M. S., Berkowitz, R. L., Callen, P. W., Canty, T. G., Catz, C., Clewell, W. H., Depp, R., Edwards, M. S., Fletcher, J.C., Frigoletto, F. D., Garrett, W. J., Johnson, M. L., Jonsen, A., de Lorimier, A. A., Liley, W. A., Mahoney, M. J., Manning, F. D., Meier, P. R., Michejda, M., Nakayama, D. K., Nelson, L., Newkirk, J. B., Pringle, K., Rodeck, C., Rosen, M.A. and Schulman, J. D.: Fetal treatment 1982. New Engl. J. Med., 307: 1651, 1982. 9. Bernstein, J.: The morphogenesis of renal parenchymal maldevelopment (renal dysplasia). Ped. Clin. N. Amer., 18: 395, 1971. 10. McCrory, W. W.: Developmental Nephrology. Cambridge, Massachusetts: Harvard University Press, 1972. 11. Beck, A. D.: The effect of intra-uterine urinary obstruction upon the development of the fetal kidney. J. Urol., 105: 784, 1971. 12. Symchych, P. S. and Winchester, P.: Potter's syndrome. Animal model: amniotic fluid deficiency and fetal lung growth in the rat. Amer. J. Path., 90: 779, 1978. 13. Harrison, M. R., Nakayama, D. K., Noall, R. and de Lorimier, A. A.: Correction of congenital hydronephrosis in utero II. Decompression reverses the effects of obstruction on the fetal lung and urinary tract. J. Ped. Surg., 17: 965, 1982. 14. Harrison, M. R., Ross, N., Noall, R. and de Lorimier, A. A.: Correction of congenital hydronephrosis in utero I. The model: fetal urethral obstruction produces hydronephrosis and pulmonary hypoplasia in fetal lambs. J. Ped. Surg., 18: 247, 1983. 15. Duckett, J. W., Harrison, M. R., de Lorimier, A. A., Arant, B. S., Jr., Bellinger, M. F., Kroovand, R. L. and Bernstein, J.: Fetal intervention for obstructive uropathy. Dial. Ped. Urol., 5: 1, 1982. 16. Inselman, L. S. and Mellins, R. B.: Growth and development of the lung. J. Ped., 98: 1, 1981. 17. Wladimiroff, J. W.: Effect of frusemide on fetal urine production. Brit. J. Obst. Gynaec., 82: 221, 1975.
Copyright© 1984 by The 1/iliiliams & ~Nilkir.;; Co.
HYDRONEPHROSIS: CONSIDERATIOl\JS REGARDING UROLOGICAL INTERVENTION DA vm A. DIArVIOND, ROGER SANDERS AND ROBERT D. JEFFS From the James Buchanan Brady Urological Institute, and Departments of Urology and Radiology, The Johns Hopkins Hospital, Baltimore, Maryland
ABSTRACT
Real-time ultrasonic imaging during pregnancy has made it possible to diagnose accurately fetal hydronephrosis as early as at 17 weeks of gestation. The ultrasonic finding of oligohydramnios in conjunction with megacystis and biiateral hydronephrosis is the most reliable indicator of marked permanent urinary tract obstruction and, therefore, is an absolute requirement for fetal urological intervention. Fetal intervention by means of percutaneous catheter drainage of the bladder into the amniotic space is undertaken to preserve renal and pulmonary function. It appears that urinary tract decompression at 17 weeks of gestation is unable to prevent renal dysplasia. The degree to which renal cortical function is preserved by fetal drainage of the hydronephrotic kidney is unknown" Experimental evidence has suggested that early urinary tract decompression promotes improved pulmonary development, which may allow for normal pulmonary function postnatally. We report 6 cases that illustrate important considerations with regard to urological intervention for fetal hydronephrosis. Real-time ultrasonic imaging during pregnancy has made it to diagnose accurately fetal lesions of the central nervous system (obstructive hydrocephalus and myelomeningocele),1 gastrointestinal system (omphalocele and diaphragmatic hernia)2 and urinary tract (hydronephrosis, hydroureter, megacystis, multicystic kidney and polycystic kidney), 3• 4 as well as detecting oligohydramnios, As noted by Kramer,5 of primary concern to the urologist is the diagnosis of bilateral hydronephrosis in utero, which has been made as early as at l 7 to 22 weeks of gestation. a, 7 In an effort to preserve renal function prenatal drainage has been achieved by hysterotomy and percutaneously, As the number of treated patients with fetal hydronephrosis has grown the advantages, limitations and disadvantages of this form of intervention have become somewhat more clear, As Harrison and associates have stated, "The most difficult problem in prenatal management is how to select only the fetuses that might benefit from treatment . , . , Careful serial assessment of fetal anatomy and organ function is necessary to exclude those so severely affected that they cannot be saved and those so mildly affected that they will do well with standard treatment," 8 We report 6 Cases in which fetal urlT'/"HlPT)hI'(';Rl CH
intervention is ultin1ately unnecessary and in which postnatal treatment produces the desired result, Case 2. B. B. VV, underwent maternal ultrasound at 34 weeks of gestation" The study demonstrated bilateral hydronephrosis, a dilated bladder and oligohydramnios (fig, 2). The fetus proceeded to term and died at birth. Autopsy findings were consistent with the prune belly syndrome, He had a markedly stenotic urethra with significant dilatation of the bladder and distal ureters. The right kidney was multicystic and dysplastic, and the left was rudimentary with a cystic dysplastic cortex (fig, 3, A), Severe pulmonary hypoplasia also was found (fig, 3, B). Comment: This case represents the most severe form of fetal hydronephrosis in which early intervention would likely have been futile. Case 30 Maternal ultrasound late in the second trimester demonstrated bilateral hydronephrosis (fig. Subsequently, monthly ultrasound examinations revealed spontaneous resoUltrasound and an IVP vvere lution of the hv,ctr<)WcPi'.11' normal after bil"tho Comment: This case demonstrates that n-n,,,r,nrn~, may be WU~UWL~ufrom underwent maternal ultrasound at 28 demonstrated bilateral Serial studies the presence of upper tract without oligohydramnios, After delivery cesarean section, abdominal ultrasound con firmed the presence of bilateral hydronephrosis and hydroureter but demonstrated a normal-appearing bladder. A voiding cystourethrogram when the newborn was 2 days old showed a normal-appearing bladder and posterior urethra with no evidence of reflux. An IVP when the neonate was 5 days old demonstrated bilateral hydronephrosis and hydrometer. Bilateral Whitaker tests showed no evidence of functional obstruction. Cystoscopy revealed a normal-appearing urethra and bladder. Comment: This case appears to demonstrate the presence of persistent, lower tract obstruction in utero, which was not present postnatally, Management was conservative, with continued surveillance of renal function and upper tract drainage, ~,,.-u<-U>~H
Case 1. M. S, had normal maternal ultrasound in the second trimester, A followup ultrasound at 36 weeks of gestation demonstrated bilateral hydronephrosis that was greater on the right side. An excretory urogram 2 days after birth demonstrated marked right hydronephrosis with mild left ureteropelvic junction obstruction. A diethylenetriaminepentaacetic acid renal scan demonstrated 3 per cent function of the right kidney, Exploration when the newborn was 7 days old revealed a multicystic kidney with an atretic ureteropelvic junction (fig. 1, A). Pathological analysis demonstrated renal dysplasia (fig. 1, B). Comment: This case is representative of those in which fetal Accepted for publication February 7, 1984.
1155
1156
DIAMOND, SANDERS AND JEFFS
FIG. 1. Case 1. A, multicystic dysplastic kidney removed when patient was 1 week old. Multiple cysts make ultrasound differentiation from ureteropelvic junction obstruction difficult. B, histologic specimen from part A shows renal dysplasia with primitive tubules and islands of loosely woven mesenchymal tissue.
FIG. 2. Case 2. Maternal ultrasound at 34 weeks of gestation demonstrates dilatation of fetal posterior urethra (PU), bladder (B) and ureter ( U).
Case 5. B. B. S. underwent maternal ultrasound at 32 weeks of gestation. The study demonstrated oligohydramnios, a dilated bladder and bilateral hydronephrosis. At 34 weeks of gestation a double-pigtailed catheter was placed percutaneously in the fetal bladder, providing urinary drainage into the amniotic cavity. The findings at birth were consistent with severe prune belly syndrome (fig. 6). Bilateral ureterostomy was performed in the neonatal period. When the infant was 4 months old renal function was normal but pulmonary function was markedly compromised secondary to reduced tidal volume, which is consistent with the prune belly syndrome. Case 6. B. B. B. underwent maternal ultrasound at 18 weeks of gestation. The study demonstrated a dilated posterior urethra, megacystis, minimal right hydronephrosis and normal amniotic fluid volume. Percutaneous aspiration of the fetal bladder was performed and subsequent bladder filling, suggesting a normal rate of urine production, was noted on ultrasound.
FIG. 3. Case 2. A, histologic specimen of right kidney shows rudimentary glomeruli and tubules in cystic dysplastic cortex. B, histologic specimen of lung reveals rudimentary alveolar pattern with leukocytic infiltrate consistent with pulmonary hypoplasia.
Followup examination at 20 weeks of gestation demonstrated the development of left hydronephrosis in addition to the previous findings. The fetal bladder was again aspirated. At 28 weeks of gestation oligohydramnios was noted on maternal ultrasound, yet at 30 weeks the fetal bladder was of normal size with no evidence of hydronephrosis on ultrasound. The newborn was delivered at 36 weeks of gestation and required
EE7 DISCUSSION
he was 16 hours demurethral valves with marked dilatation of the ~r,~,-,,~,,,w urethra, bladder and ureters, complete bilateral cystic dysplasia of the 7, A) and hyaline membrane 7, disease of the lungs Comment: In this case an indwelling percutaneous catheter was not placed because of the development of urinary ascites at the time of initial bladder aspiration, as well as unfavorable placental positioning. It is of great interest that a dilated bladder and bilateral hydronephrosis were noted on ultrasound 10 weeks before the development of oligohydramnios. It also is noteworthy that the bladder appeared to fill at a rate indicative of normal urine production at 18 weeks of gestation and yet autopsy demonstrated complete cystic dysplasia of b~th kidneys. The ultrasound findings of normal-sized bladder and kidneys at 30 weeks of gestation after long-term dilatation suggested marked deterioration in renal function at that
FIG. 4. Case 3. Maternal ultrasound during second trimester demonstrates bilateral hydronephrosis (H) that resolved spontaneously.
The value of surgery for hvdroneDhros1s tests on our to demonstrate greater tion perhaps uuu,,v,w," intervention in utero rather than in the neonatal period. Given the frequency with which renal dysplasia is found in conjunction with obstructed renal units, 9 the preservation of renal function prenatally depends upon the prevention of renal dysplasia and cortical damage due to hydronephrosis in utero. The timing of potential fetal intervention during renal embryogenesis is critical. In the human the ureteral bud appears at 5 weeks of gestation. At 9 weeks Chwalle's membrane opens and fetal urine is first detected. At 17 weeks of gestation the
FIG. 6. Case 5. Neonate with absent abdominal musculature consistent with prune be\ly ~yndro~1e. 9oubltpi~ai!ed catheter placed percutaneously to dram ,eta! unne irom o!adaer to ammotic flmd compartment is noted in lower midline.
FIG. 5. Case 4. Maternal ultrasound (transverse section) at 36 weeks of gestation. A, hydronephrosis of kidneys. B, dilated bladder and distal ureter ( U).
1158
DIAMOND, SANDERS AND JEFFS
FIG. 7. Case 6. A, histologic specimen of cystic dysplastic kidney shows rudimentary tubules surrounded by abundance of mesenchymal tissue. B, histologic specimen of pulmonary parenchyma reveals findings consistent with hyaline membrane disease.
diagnosis of hydronephrosis is possible by ultrasound. The last division of the ureteral bud is completed at 20 weeks. Between 26 and 40 weeks of gestation the induction of nephrons occurs, and 80 to 90 per cent of the renal mass develops. At term there is a sudden increase in fetal renal blood flow secondary to a decrease in vascular resistance. 10 In 1971 Bernstein noted that urinary obstruction at the onset of differentiation of the mesonephric blastema is associated with renal dysplasia. 9 He found that unilateral ureteral anomalies are associated with ipsilateral dysplasia, lower tract obstruction is associated with bilateral dysplasia and the degree of dysplasia corresponds roughly to the severity of obstruction. In 1971 Beck demonstrated that the effect of obstruction on subsequent renal development is dependent upon the gestational age of the fetus at the onset of obstruction.11 He noted that ureteral obstruction induced at 50 to 70 days of gestation in the fetal lamb (147 days of total gestation) produced renal dysplasia, whereas obstruction during the second half of gestation produced hydronephrosis but not dysplasia (fig. 8). Roughly extrapolating from Beck's work, prevention of renal dysplasia in the human would require prenatal drainage before 14 to 20 weeks of gestation. It is only towards the end of this period that hydronephrosis can be diagnosed by ultrasound. If the present diagnostic techniques do not allow intervention at a time when renal dysplasia might be prevented, one wonders if the prenatal treatment of hydronephrosis might significantly preserve renal function. Because 80 to 90 per cent of the renal mass develops during the third trimester, one might anticipate impairment of nephron formation due to back pressure from urinary obstruction during this period. While longstanding fetal hydronephrosis undoubtedly causes thinning of the renal parenchyma, the length of time that the fetal kidney can be obstructed before irreversible damage occurs has not been demonstrated experimentally. Whether the drained fetal hydronephrotic system will undergo post-obstructive diuresis is unknown, as are the cardiovascular consequences of such a response to drainage. Pulmonary hypoplasia often precludes survival in the fetus with severe obstructive uropathy. It is associated commonly with oligohydramnios, which can be diagnosed by fetal ultrasound. The exact relationship between oligohydramnios and pulmonary hypoplasia remains controversial. It is known that amniotic fluid volume and composition are determined primarily by urine production, although pulmonary secretions and secretions of the amnion itself contribute to the total amniotic
I.O
O.S
FIRST TRIMESTER
SECOND TRIMESTER
OBSTRUCTION RESULTS IN
~
THIRD TRIMESTER
OBSTRUCTION RESULTS IN HYDRONEPHROSIS
0.6 URINE PRODUCTION 0.4
0.2
20
40
60
80
100
120
140
GESTATIONAL AGE (DAYS)
FIG. 8. Variation in urine production (ml. per minute) and pathological results of urinary tract obstruction with gestational age in fetal lamb model. 10•11
fluid. Using Sprague-Dawley rats, Symchych and Winchester demonstrated that experimental deficiency of amniotic fluid in the third trimester of pregnancy was associated with pulmonary hypoplasia. 12 Harrison and associates produced experimentally urethral obstruction in fetal lambs at 100 days of gestation with resultant hydronephrosis and pulmonary hypoplasia. 13 These authors suggest that oligohydramnios results in a small intrauterine cavity that restricts the fetal thorax and prevents expansion of the fetal lung during inspiration, resulting in pulmonary hypoplasia. In half of the lambs the urinary tract was decompressed at approximately 120 days of gestation. These animals were noted to have larger lungs that were adequate to support postnatal life, in contrast to the small lungs resulting from continued obstruction until term. Whereas the developmental insult resulting in pulmonary hypoplasia in association with congenital diaphragmatic hernia and congenital hydronephrosis appears to occur early in gestation, the experiments of Harrison and associates demonstrate that the decompressed lung late in gestation is capable of resuming normal growth and development during the last trimester when total lung growth is greatest. They advocate percutaneous drainage of urine into the amniotic fluid prenatally in the selected patient to facilitate normal pulmonary development.14, 15 Primary pulmonary hypoplasia without obvious renal, thoracic or cardiac anomalies has been observed but the volume of
1159
FETAL HYDRONEPHROSIS
amniotic fluid at delivery was not described, suggesting the possibility of an amniotic fluid leak. 16 The ultrasound finding of oligohydramnios in association with obstructive uropathy appears to be of further importance. Sanders and Graham have noted 2 cases of moderately severe hydronephrosis on serial fetal ultrasound, which resolved spontaneously, as in our case 3, with normal radiographic studies at birth. 7 In neither case was oligohydramnios present. The cause of such physiological hydronephrosis is unclear. In the fetal lamb model urine production is detected first at the beginning of the second trimester, when urinary obstruction appears to produce renal dysplasia. Fetal urine production increases to a peak early in the third trimester, subsequently decreasing before birth (fig. 8). 10 Should such a pattern of urine production appear in a human, it might account for the transient hydronephrosis noted by Sanders and Graham. Therefore, we suggest that fetal drainage be contemplated only in those cases in which oligohydramnios is present, thus, making it unlikely that the hydronephrosis is transient in nature. When a fetus has bilateral hydronephrosis and oligohydramnios one would like to assess the degree of renal function remaining before fetal drainage, with its attendant risks to the mother and child. To date, there are no good means of assessing renal function in utero. Investigators have examined bladder filling and emptying using ultrasound and have observed, with ultrasound, the response in fetal urine production to furosemide administered to the mother. 17 These methods of assessing fetal renal function fail in 2 respects. Urine production is not indicative of glomerular filtration rate and can be quite high in the adult with severe cortical loss. Indeed, in our case 6 the kidneys that underwent complete cystic dysplasia appeared to have a normal rate of urine production at 18 weeks of gestation. In addition, many of the newborns with posterior urethral valves and the prune belly syndrome have massive reflux that would distort any in utero measurement of urine production based on bladder filling and emptying. Aspiration of fetal urine is technically easy but it has not yet proved to be valuable in assessing renal function in utero. One would hope that advances in this area would provide us with a more precise means of assessing renal response to in utero drainage. The assessment of potential gain by in utero drainage becomes more critical as complications of the procedure accumulate. Those reported to date include the induction of premature labor, fetal bowel perforation, and hemorrhage and infection in the fetus and maternal uterus. Harrison and associates have presented a rational schema for managing the fetus with hydronephrosis. 8 Apparent unilateral hydronephrosis (secondary to ureteropelvic junction obstruction, multicystic kidney and primary megaureter) and bilateral hydronephrosis without oligohydramnios do not require in utero treatment. We hasten to add that in cases of unilateral hydronephrosis the presence of a contralateral, normal renal unit should be established. If bilateral hydronephrosis and megacystis are accompanied by marked oligohydramnios, decompression of the urinary tract and restoration of normal amniotic fluid dynamics should be undertaken. If the fetus is too immature for delivery and postnatal therapy, percutaneous catheter decompression of the bladder into the amniotic fluid should be performed. The presence of oligohydramnios is an absolute requirement for intervention, since it is the most reliable indicator of marked permanent obstruction. However, the more severe the oligohydramnios, the worse the prognosis for recovery of renal and pulmonary function. In cases in which oligohydramnios appears questionable in nature serial ultrasound should be performed
to assess amniotic fluid volume. Immediately upon birth these newborns should be evaluated thoroughly urologically to undergo appropriate corrective procedures. At present, prenatal intervention seems most justified when marked bilateral hydronephrosis presents with oligohydramnios. Under these circumstances percutaneous drainage of the urinary tract restores amniotic fluid dynamics and attempts to preserve renal and pulmonary function. The success of fetal intervention for hydronephrosis relies upon our improved un derstanding of fetal renal physiology as well as the pathogenesis of renal dysplasia, fetal hydronephrosis and pulmonary hypoplasia. In addition, we must demonstrate experimentally that in utero drainage is of significant benefit to embryogenesis of the kidneys and lungs. REFERENCES 1. Hidalgo, H., Bowie, J., Rosenberg, E. R., Ram, P. C., Ford, K. and Lipsit, E.: In utero sonographic diagnosis of fetal cerebral anomalies. Amer. J. Rad., 139: 143, 1982. 2. Jassani, M. N., Gauderer, M. W. L., Fanaroff, A. A., Fletcher, B. and Merkatz, I. R.: A perinatal approach to the diagnosis and management of gastrointestinal malformations. Obst. Gynec., 59: 33, 1982. 3. Hadlock, F. P., Deter, R. L., Carpenter, R., Gonzalez, E. T. and Park, S. K.: Sonography of fetal urinary tract anomalies. Amer. J. Rad., 137: 261, 1981. 4. Chinn, D. H. and Filly, R. A.: Ultrasound diagnosis of fetal genitourinary tract anomalies. Urol. Rad., 4: 115, 1982. 5. Kramer, S. A.: Current status of fetal intervention for congenital hydronephrosis. J. Urol., 130: 641, 1983. 6. Garrett, W. J., Kossoff, G. and Osborne, R. A.: The diagnosis of fetal hydronephrosis, megaureter and urethral obstruction by ultrasonic echography. Brit. J. Obst. Gynaec., 82: 115, 1975. 7. Sanders, R. and Graham, D.: Twelve cases of hydronephrosis in utero diagnosed by ultrasonography. J. Ultrasound Med., 1: 341, 1982. 8. Harrison, M. R., Filly, R. A., Golbus, M. S., Berkowitz, R. L., Callen, P. W., Canty, T. G., Catz, C., Clewell, W. H., Depp, R., Edwards, M. S., Fletcher, J.C., Frigoletto, F. D., Garrett, W. J., Johnson, M. L., Jonsen, A., de Lorimier, A. A., Liley, W. A., Mahoney, M. J., Manning, F. D., Meier, P. R., Michejda, M., Nakayama, D. K., Nelson, L., Newkirk, J. B., Pringle, K., Rodeck, C., Rosen, M.A. and Schulman, J. D.: Fetal treatment 1982. New Engl. J. Med., 307: 1651, 1982. 9. Bernstein, J.: The morphogenesis of renal parenchymal maldevelopment (renal dysplasia). Ped. Clin. N. Amer., 18: 395, 1971. 10. McCrory, W. W.: Developmental Nephrology. Cambridge, Massachusetts: Harvard University Press, 1972. 11. Beck, A. D.: The effect of intra-uterine urinary obstruction upon the development of the fetal kidney. J. Urol., 105: 784, 1971. 12. Symchych, P. S. and Winchester, P.: Potter's syndrome. Animal model: amniotic fluid deficiency and fetal lung growth in the rat. Amer. J. Path., 90: 779, 1978. 13. Harrison, M. R., Nakayama, D. K., Noall, R. and de Lorimier, A. A.: Correction of congenital hydronephrosis in utero II. Decompression reverses the effects of obstruction on the fetal lung and urinary tract. J. Ped. Surg., 17: 965, 1982. 14. Harrison, M. R., Ross, N., Noall, R. and de Lorimier, A. A.: Correction of congenital hydronephrosis in utero I. The model: fetal urethral obstruction produces hydronephrosis and pulmonary hypoplasia in fetal lambs. J. Ped. Surg., 18: 247, 1983. 15. Duckett, J. W., Harrison, M. R., de Lorimier, A. A., Arant, B. S., Jr., Bellinger, M. F., Kroovand, R. L. and Bernstein, J.: Fetal intervention for obstructive uropathy. Dial. Ped. Urol., 5: 1, 1982. 16. Inselman, L. S. and Mellins, R. B.: Growth and development of the lung. J. Ped., 98: 1, 1981. 17. Wladimiroff, J. W.: Effect of frusemide on fetal urine production. Brit. J. Obst. Gynaec., 82: 221, 1975.