Pediatric Nephrology
Pediatr Nephrol (1988)2:334-342 9 IPNA 1988
Invited review
Prenatal diagnosis and management of bilateral hydronephrosis Timothy M. Crombleholme, Michael R. Harrison, Michael T. Longaker, and Jacob C. Langer Department of Surgery (Division of Pediatric Surgery), University of California, San Francisco, 3rd and Parnassus Avenues, HSE 585, San Francisco, CA 94143-0510, USA
Abstract. This report reviews the management of the fetus with congenital hydronephrosis (CH), a challenging diagnostic and therapeutic problem. Experimental models of obstructive uropathy have produced histologic changes similar to those seen in kidneys of human neonates with congenital hydronephrosis. Relief of obstruction in utero in these models has been shown to prevent some of the dysplastic changes caused by obstruction. These studies have formed the theoretical basis for in utero decompression to restore amniotic fluid dynamics to prevent death from pulmonary hypoplasia, and reverse or arrest dysplastic morphogenesis. The development of prognostic criteria has greatly aided in selection of appropriate fetuses for intervention. These criteria include: (1) Na < 100 mEq/1; (2) C1 < 90 mEq/1; (3) osmolarity <210 mosmol; (4) sonographic appearance of the fetal kidneys; (5) amniotic fluid status; (6) urine output at fetal bladder catheterization. All fetuses should have ultrasonography to exclude other anomalies, and karyotype analysis to exclude chromosomal abnormality. If amniotic fluid volume is normal, the pregnancy is followed with serial ultrasound examinations. If oligohydramnios develops, a prognostic evaluation is performed, including fetal bladder catheterization. If the fetus has poor residual renal function, on the basis of prognostic criteria, appropriate counseling may be given. If the fetus has good residual renal function, depending on lung maturity, it can be delivered early for corrective surgery. If diagnosed prior to lung maturity in utero, decompression by either vesicoamniotic shunting or
Offprint requests to: M. R. Harrison
open fetal surgery may be attempted in the highly selected case. Very few fetuses with CH will require in utero decompression, but all benefit from early diagnosis allowing preparation for postnatal care. Open fetal surgery should be considered an experimental therapy until efficacy and safety are established in controlled trials. Key words: Congenital hydronephrosis surgery - Vesicoamniotic shunts
Fetal
Introduction The fetus with congenital hydronephrosis presents a difficult diagnostic and therapeutic challenge. The clinical management requires both an accurate delineation of the abnormality and a thorough understanding of the pathophysiology of obstruction and its sequelae in the developing fetus [1-3]. Fortunately, obstetric ultrasonography has enabled clinicians to diagnose fetal hydronephrosis early enough in pregnancy to begin defining the natural history of fetal urinary tract obstruction. The pathophysiology of fetal hydronephrosis has been studied in the fetal lamb model, and the feasibility and safety of intervention has been studied in a rigorous primate model [4-6]. The rationale for early decompression of the obstructed urinary tract is straightforward: unrelieved obstruction causes progressive damage to the developing kidney and lungs, which compromises survival at birth [7]. Although we have been able to demonstrate experimentally that early decompression can ameliorate the development of renal dysplasia and pulmonary hypoplasia, efficacy in human fetal hydronephrosis remains unproven.
335 The most challenging clinical problem is how to select from the large number of fetuses with dilated urinary tracts those few for whom intervention is appropriate, that is, those with obstruction severe enough to compromise renal and pulmonary function at birth, but not so severe that damage is irreversible. Experimental work and a growing clinical experience has contributed to improved management of the fetus with obstructive uropathy. Counseling of the family and medical management of the fetus is now based on a more thorough knowledge of the natural history and pathophysiology of the disease, as well as quantitative and qualitative tests of renal function. Natural history of obstruction before and after birth
Knowledge of the natural history of fetal urinary tract dilation is essential for clinically appropriate evaluation and management. Although the natural history and outcome of urinary tract obstructions after birth are well documented, we are only beginning to document the natural history of fetal obstruction. We are finding that the course and outcome are very different for fetuses diagnosed before birth because the most severely affected fetuses have not survived long enough after birth to come to the attention of pediatric nephrologists. The incidence of hydronephrosis present at birth ranges from 1/200 to 1/1000 births, with severe forms occurring less frequently. Williams et al. [8] reviewed 31 infants with posterior urethral valves (PUV) who were less than 1 month old when first seen; whilst 16 died (52%), none did so
from respiratory insufficiency. In a subsequent report of 100 cases of PUV presenting at 2 weeks or less, only 1 of 24 infants died (4%) [9]. No deaths were reported by Evans and Lorenzo [10] in 10 cases of PUV aged 1 year or less. All of these studies reflect a selection bias. These infants had less severe forms of the disease and thus had no respiratory complications. To study this issue, Nakayama et al. [11] reviewed 11 cases of PUV that were clinically evident at birth; 5 of 11 infants died (45%). Three died within 2 h of birth of respiratory insufficiency, and 2 succumbed within 3 weeks to rapidly progressive renal failure. Of the 6 survivors, 4 had respiratory problems including prolonged ventilatory support (2) and bilateral pneumothoraces (2). Only 2 of the 11 infants had normal renal function after surgical relief of obstruction. Fetal urethral obstruction, usually due to PUV in males, produces a wide variety of clinically significant effects. Oligohydramnios from lack of fetal urine output may result in pulmonary hypoplasia, Potter facies, and limb abnormalities such as club foot or hip dislocation. Abdominal distention by urinary ascites or massive dilation in early gestation (often with spontaneous decompression later in gestation) results in the prune belly deformation [12, 13] (Fig. 1). Oligohydramnios secondary to obstruction and decreased fetal urine output produces pulmonary hypoplasia. If oligohydramnios occurs early (<24 weeks) and persists, postnatal respiratory insufficiency and death can be anticipated. In babies not dying at birth from pulmonary hypoplasia, progression to end-stage renal disease despite
CONSEQUENCES OF URETHRAL OBSTRUCTION
.72--\" .
\ \
~,OHYDRAMNIOS
i\\ II I I ; ! fDRONEPHROSIS / / II
, I
/ YDROURETER g MEGACYSTIS ~
/
POTTER'S FACI HYPOPLASTIC LUNGS FLEXION CONTRACTIOI, Ii, TYPE IV CYSTI DYSPLASIA RENAL
FAILURE ABDOMINAL MUSCLE DEFICIENCY SYNDROME " PRUNE BELLY"
1 Fig. 1. Consequences of urethral obstruction
336 early postnatal urinary diversion implies that kidney damage occurred prenatally and was not reversible after birth. The rationale for diagnosis and treatment of a few highly selected cases in utero is that life-threatening damage to lungs and kidneys may be averted or ameliorated by early decompression (Fig. 1).
Experimental pathophysiology of fetal obstruction The development of experimental lamb and primate models has begun to elucidate the pathophysiology of fetal urinary tract obstruction. Experimental studies suggest that the severity of damage present at birth appears to depend on the type, degree and duration of obstruction [1, 2, 4, 5, 14]. Beck [15] performed ureteral ligation as well as contralateral nephrectomy in lambs from 62 to 84 days' gestation. The effect of ureteral ligation
(•
Group 1
depended upon the gestational age of the fetus at onset of obstruction. Hydronephrosis alone resulted when the ureter was ligated during the last half of gestation. When ligation was performed during the first half of gestation, the kidneys were small and contracted, resembling human dysplastic kidneys. We produced unilateral obstruction in fetal lambs at 55-65 d a y s ' gestation (term 140-145 days) [161. At term, the affected kidney was both hydronephrotic, with ureteral and calyceal dilation, and dysplastic (Fig. 2). Cut sections showed capsular scarring and numerous cortical cysts. Microscopy of the atrophic obstructed kidney showed diffuse fibrosis, especially in the peripelvic area, which extended into the renal interstitium and around large vessels. Numerous dilated tubules and ectatic lymphatics, which formed subcapsular microcysts, were present. These changes, in particular the fibrosis, distorted
(~1
Group 5A
(~
N=3
(~(~
N=]
a
a (~
Group 2
IIIIIII
(~
(~ Group SB
N=2 N=I
b
iiiJii]iiii!!!!!!i
Group 4
N=2
N-3
C
c 0
20
40
I00 120 140
0
Term
GESTATION (days)
2
::::::::::::::::::::::::
DEVELOPMENT: [ ~
NORMAL ~
20
40
6
8
100 120 140
GESTATION (days)
OBSTRUCTED ~
DECOMPRESSED
3
DEVELOPMENT:~
=
Term
NORMAL ~ ]
OBSTRUCTED
Fig. 2 a - e . Fetal lamb renal dysplasia model, histopathological changes. Histology of kidneys obstructed for different periods of time prior to decompression in utero, a Kidney from group 1 lamb (obstructed 65 days, decompressed 85 days) shows minimal cortical changes and minimal interstitial fibrosis associated with small subcapsular dilated structures. There is atrophy of medullary pyramids with loose pericalyceal fibrosis (hematoxylin-eosin; x 6.5). b Kidney from group 2 lamb (obstructed 65 days, decompressed 105 days) shows cortical features similar to group 1, but slightly increased pericalyceal fibrosis (hematoxylin-eosin; x 6.5). e Kidney from group 3 lamb (obstructed 65 days, decompression 125 days) shows cortical features similar to a and b, but increased pericalyceal and corticomedullary pyramid atrophy. There is a marked histologic difference between this group and group 5B (see Fig. 3) (hematoxylin-eosin; x 5.5). (Reprinted with permission [77]) Fig. 3 a - e . Histology of kidneys still obstructed at time of killing, a Kidney from group 5A lamb (obstructed 65 days, sacrificed 85 days) shows dilatation of subcapsular structures with some dilated medullary tubules. There is increased interstitial fibrosis tissue throughout, especially in pericalyceal area (hematoxylin-eosin; x 5.5). b Kidney from group 5B lamb (obstructed 65 days, sacrificed 125 days) shows severe interstitial fibrosis in cortex, severe calyceal fibrosis, marked medullary pyramid atrophy, and few dilated subcortical structures. Although distorted by fibrosis, the changes are insufficient for dysplasia (hematoxylin-eosin; x 5.5). e Kidney from group 4 lamb (obstructed 65 days, delivered 140 days) shows parenchymal atrophy and severe fibrosis throughout. Resulting disorganization is consistent with renal dysplasia (hematoxylin-eosin; x 5.5). (Reprinted with permission [77])
337 the cortex and medulla producing focal, parenchymal disorganization. The unobstructed contralateral kidney was unaffected. These pathologic changes are strikingly similar to those seen in kidneys of human neonates with obstructive uropathy [17]. These experiments suggested that the renal dysplasia associated with obstructive uropathy may be caused by simple obstruction to the flow of fetal urine early in gestation [18, 19]. In subsequent studies, we showed that decompression of ureteral obstruction prior to term prevents renal dysplasia and produces reversible postobstructive changes (Figs. 2, 3). At term, the degree of pathologic changes seen in the obstructed and then decompressed kidneys appeared proportional to the length of time the obstruction existed. Both timing and duration of obstruction would seem to be crucial in producing dysplastic morphogenesis.
Diagnostic methods Ultrasound. Detection and anatomical delineation of fetal obstructive uropathy depend on accurate ultrasonography. In our experience and that of others [20, 21] the anatomy of most urinary tract anomalies can be accurately defined, distinguishing unilateral from bilateral disease, lower (urethral) from higher (uretero-pelvic junction, UPJ) obstructions, and distinguishing various types of parenchymal renal disease (e. g. polycystic kidney disease). Most important, significant associated malformations can usually be evaluated by ultrasound and chromosome analysis by amniocentesis, chorionic villous sampling or fetal blood sampling. The only error in sonographic assessment of the urinary tract anatomy at our institution was the inability to distinguish whether the fluid collection in UPJ obstruction was in the dilated pelvis or a paranephric collection from calyceal rupture [21]. Ultrasound provides limited information about urodynamics and renal function, but some information can be inferred from the volume of amniotic fluid and the presence or absence of urine in the bladder. Oligohydramnios is probably the most important indicator of fetal functional renal impairment [1]. Because fetal urine is a major source of amniotic fluid in late pregnancy, it is not surprising that oligohydramnios accurately reflects decreased fetal urine excretion, and that normal amniotic fluid volume accurately predicts the presence of at least one functioning kidney [22]. It is not possible to accurately quantify the exact volume of fluid, but serial ultrasound examinations can show a trend in amniotic fluid volume. The best indicator of the severity of disease appears to be the time in gestation when oligohydramnios develops. We remain pessimistic about the fetus that develops severe oligohydramnios before the 20th week of gestation [23, 24]. The fetal urinary bladder can be imaged easily at 18 menstrual weeks and often as early as 15 weeks [25]. The bladder normally increases in size and empties in a cyclical pattern. It is not uncommon for this to take place during an examination. It may be necessary to ask the patient to stay for 1-2 h for serial observation of bladder filling and emptying. Non-visualization of the fetal urinary bladder in association with oligohydramnios indicates a poor pregnancy outcome. Although abnormally dilated ureters are easily identified, non-dilated fetal ureters cannot be visualized. We have only
seen one case in which a non-dilated ureter was identifiable possibly because of vesicoureteric reflux [25]. Ultrasound evaluation of fetal kidneys has proven useful to predict severe dysplasia. The demonstration of renal cortical cysts predicts dysplasia with 44% sensitivity and 100% specificity. When renal cysts are not sonographically demonstrable, however, one cannot exclude the presence of renal dysplasia. Assessment of renal echogenecity had a sensitivity of 73% and a specificity of 80%, in predicting dysplasia. The presence of hydronephrosi s was least predictive of dysplasia (sensitivity 41%, specificity 73%), emphasizing that dilation does not necessarily imply renal damage. In the fetus, as in the child, the urinary system may remain very dilated after it has been functionally decompressed by extravasation, which often results in urinary ascites or urinoma [16, 27-29]. Mild to moderate degrees of renal dysplasia are difficult to assess sonographically.
Contrast radiography. In our early experience, there were several cases where ultrasound could not establish the level of obstruction. We used antegrade fetal pyelography with contrast placed into the renal pelvis under ultrasonic guidance [30]. Occasionally, a fetal cystogram may be useful in evaluating a dilated but not obstructed urinary tract (e. g. primary reflux). Magnetic resonance imaging. Magnetic resonance imaging (MRI) is a recently developed non-invasive imaging technique that produces high-resolution cross-sectional body images [31]. MRI shows promise when applied to fetal imaging. With MRI no ionizing radiation is used and, to date, there are no known adverse biologic effects [32]. Human lymphocytes, hamster ovary cells and various bacteria have been exposed in vitro without apparent side effects [33-36]. Johnson et al. [37] scanned 15 pregnant women with MRI at 10-20 weeks and saw no change in fetal heart rate or Doppler-determined umbilical artery blood flow. They also found no fetal heart rate changes after imaging in late pregnancy. The diagnostic potential of MRI in evaluating developing fetal organ systems including the brain, lungs, cardiovascular, GI and G U tracts has been documented [38-52]. MRI appears to have several advantages o v e r ultrasound, including the imaging of patients with oligohydramnios. MRI does not depend on amniotic fluid for successful visualization of the fetus and, since the fetus usually does not move very much in the absence of amniotic fluid, important information about associated anomalies (e. g. obstructive uropathy) may be provided [45]. While ultrasound will remain the primary obstetric imaging modality, M R / shows potential as a supplement to ultrasound in the diagnosis of some fetal anomalies. Fetal urine analysis. The most difficult problem with management of the fetus with hydronephrosis has been h o w t o assess renal functional potential and thus be able to select the fetus that might benefit from treatment. Sonographic analysis, as stated above, is helpful for the case of obstruction with normal amniotic fluid volume, or the other extreme of severe oligohydramnios and severe dysplasia. However, it is in the grey zone between these two extremes where potentially fatal renal and pulmonary damage may be averted by decompression. We therefore investigated the evaluation of the functional potential of the fetal urinary tract to predict postnatal renal and pulmonary function [14]. We reviewed the management of 20 fetuses with bilateral hydronephrosis. The fetuses were retrospectively divided into two groups. Group A (n = 10) had poor function due to severe renal dysplasia and pulmonary hypoplasia at autopsy or biopsy, or renal and pulmonary insufficiency at birth. Group B
338 Table 1 Group
Predicted function
Amniotic fluid status at the time of initial presentation*
I (N = 10)
Poor
II (N = 10)
Good
Ultrasound appearance of kidneys**
Fetal urine Sodiurn* (mEq/ml)
Chloride* (mEq/ml)
Potasslum*** (mEq/ml)
Creatinine*** (mg/dI)
Osmolarity* (mosmol)
Output**** (ml/H)
Iothalamate excretion (ng/H)***
Mild decrease (1) Normal (3) Moderate decrease (3) Echogenic (4) Severe decrease (6) Cystic (1) Both (2)
126• 12 (n = 8)
110~10 (n=9)
3.6• (n=9)
1.3• (n=7)
261• (n=8)
1.5• (n=5)
44• (n=2)
Normal (7) Normal (7) Mild decrease (2) Echogenic (4) Moderate decrease (1) NA (1)
67 • 16 (n = 9)
58• (n=9)
2,9• (n=9)
2.0• (n=8)
150• (n=9)
6,3• (n=4)
125• (n=4)
* P <0.001; ** P <0.05; *** P>0.05; **** P<0.02
(n = 10) had good function with non-dysplastic kidneys at autopsy or biopsy, or normal renal and pulmonary function at birth. The two groups had urine collected by percutaneous insertion of a balloon-tipped catheter (4F, Model JC-2U, Critekon, Inc., Tampa, Fla.) into the dilated fetal bladder u n d e r ultrasonographic guidance. The urinary tract was completely drained of urine and an aliquot of this urine was analyzed. The two groups were compared for: (1) amniotic fluid status at presentation; (2) ultrasound appearance of the fetal kidneys; (3) fetal urine composition [specifically concentrations of sodiu m (Na), potassium (K), chloride (C1), creatinine (Cr) a n d osmolarity (osm)]. After maternal hydration (500 ml normal saline, IV, in 1 h) and an intravenous bolus of iothalamate (0.15 ml/kg, Conroy 60, Mallenckrodt, St. Louis, Mo., USA) every hour, fetal urine output and iothalamate excretion were measured. The catheter was removed after 6 h (range 3 - 1 6 h). Fetal urine iothalamate concentrations were measured by excitation electron analysis. The results are summarized in Table 1 [14]. Amniotic fluid volume and sonographic appearance of the fetal kidneys were of prognostic significance. In addition, the sodium and chloride concentration a n d osmolarity of the fetal urine and urine output were significantly different in the fetuses defined as having good function compared with those with poor function. We found that fetuses normally excrete a hypotonic urine, whereas those with poor function had isotonic urine. We were able to establish prognostic criteria based on these results to identify those fetuses with good function and those with poor function (Table 2) [14]. In 18 fetuses (90%) the prognostic criteria were all correct, a n d d n 2 fetuses (10%), all but one of the prognostic criteria were correct. We have used these prognostic criteria to predict the potential for recovery and believe that it has been helpful in counseling the families and selecting appropriate management.
Experimental work on assessment of fetal renal function. We felt that an accurate quantitative assessment of fetal renal function would be helpful to predict the potential for recovery of fetal kidneys with obstructive uropathy. We had to consider whether endogenous fetal creatinine clearance was an accurate measure of glomerular filtration rate (GFR) in obstructed fetal kidneys. We used a fetal lamb model and compared fetal creatinine clearance with a standard test for G F R (iothalamate clearance) [53]. A substance used clinically for fetal G F R determination should satisfy the following requirements: (1) clearance exclusively by glomerular filtration; (2) freely crosses the placenta; (3) non-radioactive; (4) easily quantitated in plasma and urine with a high degree of accuracy. Theoretically, if maternal and fetal plasma levels of the substance are equal, the fetal G F R can be determined by fetal urine collection and maternal blood sampling alone, which obviates the need for fetal blood sampling. In our study, six fetal lambs underwent left ureteral ligation at 60-63 days' gestation. This procedure results in progressive renal parenchymal injury with unilateral renal dysplasia at term (145 days). At a second operation at 113-120 days' gestation, each fetus underwent placement of: (1) ureterostomy catheter to drain the obstructed kidney; (2) suprapubic bladder catheter to drain the contralateral unobstructed control kidney; (3) femoral arterial catheter for blood pressure monitoring and blood sampling; (4) femoral venous catheter for iothalamate infusion. Renal function was measured by hourly split urine collections for determination of creatinine clearance, iothalamate clearance and fractional sodium excretion from each kidney over a 4-h period. A 4-h collection period was chosen to simulate our clinical diagnostic urinary catheter experience with h u m a n fetuses with obstructive uropathy.
Table 2. Prognostic criteria for the fetus with bilateraI obstructive uropathy Predicted function
Amniotic fluid status at the time of initial presentation
Sonographic appearance of kidneys
Poor
Moderate to severely decreased
Good
Normal to moderateley decreased
Fetal urine Sodium (mEq/ml)
Chloride (mEq/ml)
Osmolarity (mosmol)
Output (ml/U)
Echogenic to cystic
> 100
>90
>210
<2
Normal to echogenic
< 100
<90
<210
>2
339 (1) rule out life-threatening anomalies; (2) confirm the diagnosis of congenital hydronephrosis; (3) delineate the anatomy of the urinary tract and the level of obstruction; and (4) define the baseline status of amniotic fluid volume. Serial ultrasound examinations will be necessary to follow the fetus to term. Extremes of amniotic fluid volume, i.e. normal versus severe oligohydramnios, has proved a reliable predictor of postnatal renal and pulmonary function and fetal outcome. When oligohydramnios develops secondary to urinary tract obstruction before 24 weeks' gestation (during the canalicular stage of lung development), the fetus will usually have renal dysplasia (type IV cystic disease) and pulmonary hypoplasia that precludes neonatal survival [3, 5, 11, 14, 21, 24, 64]. Renal parenchymal development or maldevelopment is complete at birth and, for that reason, relief of obstruction in infancy and childhood may not prevent progression to end-stage renal disease [64-66]. Likewise, hereditary types of dysplasia are unlikely to respond to decompression as they result from disturbed organogenesis prior to the 8th week of gestation. The most common type of Cystic dysplasia appears to be a developmental consequence of obstruction during nephrogenesis [67-70]. The severity of dysplasia depends on the timing and severity of obstruction. Early and complete obstruction affects tubular and glomerular differentiation throughout the kidney, resulting in irreversible dysplastic changes. The more common type of dysplasia affects cortical structures that develop later in gesta-
We found an excellent correlation between creatinine clearance and iothalamate clearance in either the obstructed or contralateral unobstructed kidneys. Creatinine clearance consistently overestimated G F R measured as did the iothalamate clearance. Compared with the control side, the obstructed fetal kidney also demonstrated abnormal tubular function with marked sodium loss as determined by either the fractional sodium excretion ( P < 0.0005) or the sodium concentration in the initial urinary aspirate ( P < 0.005). Numerous studies in children and adults have demonstrated that endogenous creatinine clearance is a reasonably accurate estimate of G F R [54-56]. There are several advantages of creatinine clearance for fetal G F R determination. First, creatinine is an endogenous, inert compound that is normally distributed in total body water, so a maternal infusion of this substance is unnecessary. Second, studies in rats, dogs, s u b h u m a n primates and humans have shown that creatinine rapidly crosses the placenta bidirectionally and that maternal and fetal plasma creatinine levels are equal [57-60]. Plasma creatinine equilibrium is maintained between the mother and the nephrectomized primate, which indicates the potential of the placenta to act as the sole fetal excretory mechanism [61]. Finally, creatinine can be easily and accurately detected in plasma and urine. Our studies have demonstrated that fetal creatinine clearance correlates well with fetal G F R measured in either normal or chronically obstructed fetal kidneys, and thus is a promising fetal renal function test. Clinical implementation will require correlation of fetal creatinine clearance with estimated fetal weight [62] or gestational age, both of which correlate strongly with increases in G F R in the fetal lamb [63]. However, in our previously mentioned study of prognostic criteria, iothalamate excretion did not appear to discriminate good function from poor function [14]. Further studies are needed to determine if fetal G F R measurement is helpful to predict ultimate postnatal renal function in man.
Clinical management The initial management of the fetus with suspected congenital hydronephrosis should include a thorough ultrasonographic examination to:
Nephrogenesis complete Continued growth / /
/
.2
I
~
1.0 i//
O.8 o ra,,,"
z
a <
•
30
,,/'
z
-r m iii
'
Collecting ducts / complete (XV) I
0 ~l -'-
JUreteric bud (I)
0.4 ~-\ First nephron J ~ Onset renal I ,~L
u.z~
~ \ \
% Mesonephros Pronephros
\\
I
1/3 Nephrons / 1/lO Moss/ l/ /
/
function / First permanent /
nephrons ( V , ) / , ,
/ .J~ ,"
/
~
2O
!
I
, ~
//
/
..J l0
< Z
UJ
,,*'
s"
2'o
,"
3'o
GESTATIONAL AGE
5 wk .
Fig. 4. Fetal renal development is presented schematically. The branching of the collecting system (15 generations) is complete by 20 weeks, but the majority of nephrons and most of the functional mass form in the cortex after 20 weeks. Obstruction interferes with nephrogenesis; relief of obstruction may allow further development. (Reprinted with permission [3])
340 tion [65-70]. Relief of obstruction during the most active phase of nephrogenesis, between 20 and 30 weeks, may obviate further damage and allow nephrogenesis to proceed (Fig. 4). The approach to the patient presenting with congenital hydronephrosis is summarized in Fig. 2 [71]. If there are no other detectable defects and the amniotic fluid volume is adequate, the pregnancy is followed by weekly ultrasound. This serial ultrasound approach is crucial for several reasons. Urinary tract dilation can occur transiently in up to 20% of normal pregnancies. This is thought to be due to smooth muscle relaxation under the influence of maternal hormones, and also the increased rate of fetal urine production, approximately 5 ml/kg per hour. In addition, urinary tract dilation can resolve spontaneously either because mild obstruction resolves with growth, or because dilation was due to non-obstructive mechanisms such as ureterovesical reflux [31. Serial examinations to detect decreasing amniotic fluid volume are performed. As long as good amniotic fluid volume is maintained, the pregnancy is followed to term and the congenital hydronephrosis is corrected after birth. Prenatal diagnosis clearly improves perinatal management and allows for earlier correction of congenital hydronephrosis after birth. This may prevent further renal damage and potentially salvages the kidney in an otherwise clinically undetectable disease [3]. Those families with pregnancies found to have associated life-threatening anomalies can be counseled regarding termination. Those pregnan-
cies which present with oligohydramnios or develop diminishing amniotic fluid during serial follow-up examinations should undergo a prognostic evaluation. This evaluation includes: (1) assessment of amniotic fluid status; (2) assessment of sonographic appearance of renal parenchyma; (3) temporary bladder catheterization to determine urine Na, C1, osmolarity and urine output; and (4) fetal karyotype. Fetal urine is produced by the 13th week of gestation and is an ultrafiltrate of fetal serum. This ultra filtrate becomes hypotonic by the selective tubular absorption of N a + C 1 - ions in excess of free water [72-74]. As stated previously, if fetal urine is hypotonic, it predicts intact glomerular and tubular function as well as good fetal outcome (Table 1) [14, 75, 76]. Fetal urine output was found to be an accurate predictor of fetal renal function [141. Fetal urine output, however, has several inherent potential problems including the state of maternal hydration, incomplete drainage of the upper tracts, the potential for postoperative diuresis in dysplastic kidneys and the very small volume of urine used when discriminating poor from good function. Those fetuses with clearly compromised function, as indicated by cortical cysts on ultrasound, oligohydramnios, and high Na, CI, osmolarity and isotonic urine, represent a subgroup with irreversible disease and will not benefit from intervention. Those fetuses who are found to have good prognosis for recovery of renal function if decompressed and are of sufficient gestational age to have pulmonary maturity should have early deliv-
THE FETUS WITH BILATERAL HYDRONEPHROSI$
1
IULT.ASOUNDEVALUAT, NO OTHER DETECTABLE DEFECTS
OTHER LIFE TREATENING DEFECTS
ADEQUATE AMNIOTIC FLUID
OLIGOHYDRAMNIOS
.oOTiC
I ULTRASOUNDAND1 / COUNSELWEEKLYJ
~a~-~'o
R ~ I''- AF S..... ~ " ~li' " " / Rlmo',chodensiiy X / { TemporQryBlodderCcilhe~er ~l~Jl~lDECREASING AMNIOTIC
\
&2 g;';Tm
POOR
/
AMNIOTIC
FLUID STA LE
GOOD
FUNCTION
FUNCTION
/
LUNGS MATURE
\
LUNGS IMMATURE
1 NO INTERVENTION
TERMINATE PREGNANCY
DELIVER EARLY" ~ / ~ O u M P R R E O ~ DELIVERAT TERM CORRECT EX U T E R o ~ C O R R E C T EX UTERO
Fig. 5. The fetus with bilateral hydronephrosis
341 cry for immediate ex vivo repair. Prenatal interventions should be reserved for those fetuses with adequate renal function for postnatal survival and pulmonary immaturity precluding early delivery. The two options for intervention are percutaneous catheter placement and fetal surgery. The percutaneous catheter technique has been successful but require s close observation as it is prone to obstruction, migration and dislodgement by the fetus, necessitating frequent replacement. We currently favor use of percutaneously placed shunts for short-term decompression late in gestation while awaiting lung maturity [3]. Open fetal surgery is favored when weeks to months of decompression will be required until fetal maturity is achieved. Management of the fetus with hydronephrosis has improved dramatically in the last few years (Fig. 5). Prognosis can now be realistically assessed and families counseled appropriately [3, 14, 16, 21, 25, 64]. Many fetuses can be identified as irreversibly damaged and inappropriate intervention can be avoided; the family may choose termination of the pregnancy. Most fetuses with obstructive uropathy can be successfully managed after birth; prenatal diagnosis allows the mother to be moved to an appropriate hospital and improved postnatal care. Very few fetuses will benefit from prenatal intervention but, if carefully selected, decompression by vesicoamniotic catheters or surgical exteriorization appears feasible and reasonably safe in experienced hands. However, surgical treatment of fetal hydronephrosis should be considered experimental until efficacy and safety are established in controlled studies. References 1. Harrison MR, Filly RA, Parer JT (1981) Management of the fetus with a urinary tract malformation. JAMA 246: 635-639 2. Harrison MR, Golbus MS, Filly RA (1981) Management of the fetus with a correctable congenital defect. JAMA 246:774-777 3. Harrison MR, Golbus MS, Filly RA (1982) Management of the fetus with congenital hydronephrosis. J Pediatr Surg 17:728-742 4. Harrison MR, Ross NA, Noall R (1983) Correction of congenital hydronephrosis in utero. I. The model: fetal urethral obstruction produces hydronephrosis and pulmonary hypoplasia in fetal lambs. J Pediatr Surg 18:247-256 5. Harrison MR, Nakayama DK, Noall R (1981) Correction of congenital hydronephrosis in utero. II. Decompression reverses the effects of obstruction on the fetal lung a~d urinary tract. J Pediatr Surg 17:965-974 6. Harrison MR, Anderson J, Rosen MA (1982) Fetal surgery in the primate. I. Anesthetic, surgical and tocolytic management to maximize fetal-neonatal survival. J Pediatr Surg 17:115-122
7. Flake AW, Adzick NS, Glick PL (1987) Evaluation of the fetus with hydronephrosis. New techniques in urology. Futura Publishing, New York, pp 269-286 8. Williams DI, Whitaker RH, Barrett TM (1983) Urethral valves. Br J Urol 45:200 9. Williams DI (1977) Urethral valves: a hundred cases with hydronephrosis. In: Bergma D, Duchett JW (eds) Urinary system malformations in children. Liss, New York, pp 342-367 10. Evans SC, Lorenzo RL (1979) Posterior urethral valves: current concepts in diagnosis and treatment. J Urol 124:76 11. Nakayama DK, Harrison MR, de Lorimer AA (1986) Prognosis of posterior urethral valves presenting at birth. J Pediatr Surg 21 : 43-45 12. Thomas IT, Smith DW (1974) Oligohydramnios: cause of non-renal features of Potter's syndrome including pulmonary hypoplasia. J Pediatr 84:811-814 13. Pagan RA, Smith DW, Shepard TH (1979) Urethral obstruction malformation complex: a cause of abdominal muscle deficiency and the "prune belly". J Pediatr 94: 900-906 14. Glick PL, Harrison MR, Golbus MS (1985) Management of the fetus with congenital hydronephrosis. II. Prognostic criteria and selection for treatment. J Pediatr Surg 20:376 15. Beck AD (1971) The effect of intrauterine urinary obstruction upon development of the fetal kidney. J Urol 105: 784-789 16. Glick PL, Harrison MR, Noall RA (1983) Correction of congenital hydronephrosis in utero. III. Early mid-trimester ureteral obstruction produces renal dysplasia. J Pediatr Surg 18:681-687 17. Potter EL, Craig JM (1975) Pathology of the fetus on the infant, 3rd edn. Year Book, Chicago , pp 458-459 18. Risdon RA (1971) Renal dysplasia. J Clin Pathol 24:57 19. Bernstein J (1978) Renal hypoplasia and dysplasia. In: Edelmann CM (ed) Pediatric kidney disease, volume II. Little Brown, Boston, pp 547-554 20. Bond SJ, Harrison MR (1987) Obstructive uropathy when to intervene. Contemp Obstet Gynecol pp 64-74 21. Golbus MS, Harrison MR, Filly RA (1983) Prenatal diagnosis and treatment of fetal hydronephrosis. Semin Perinatol 7:102-105 22. Abramovich DR (1978) The volume of amniotic fluid and its regulating factors. In: Fairweather D, Eskes T (eds) Amniotic fluid: research and Clinical Application, 2nd edn. Excerpta Medica, Amsterdam New York, pp 31-49 23. Harrison MR (1983) Progress in managing hydronephrosis. Contemp Obstet Gynecol 22:47-77 24. Harrison MR, Golbus MS, Filly RA (1982) Fetal surgery for congenital hydronephrosis. N Engl J Med 306: 591-593 25. Harrison MR, Golbus MS, Filly RA (1984) The unborn patient: prenatal diagnosis and treatment. Grune and Stratton, Orlando, pp 33-118 26. Mahoney BS, Filly RA, Callen PW (1984) Sonographic evolution of fetal renal dysplasia. Radiology 152:143 27. Glazer GM, Filly RA, Callen PW (1982) The varied sonographic appearance of the urinary tract in the fetus and newborn with urethral obstruction. Radiology 144: 563-568 28. Rottner WH, Meyer R, Bernstein J (1963) Congenital abnormalities of the urinary system. IV. Valvular obstruction of the posterior urethra. J Pediatr 63 : 84-94 29..Parker RM (1974) Neonatal urinary ascites: a potentially favourable sign in bladder outlet obstruction. Urology III: 589-593
342 30. Gore RM, Callen PW, Filly RA (1982) Prenatal percutaneous antegrade pyelography in posterior urethral valves: sonographic guidance. Am J Roentgenol 139:994 31. Low TW, Weinreb J, Santos-Ramos R (1985) Magnetic resonance imaging in human pregnancy. Obstet Gynecol 66:629 32. Budringer TJ (1981) Nuclear magnetic resonance (NMR) in vivo studies: known thresholds for health effects. J Comput Assist Sonogr 5:800 33. Thomas A, Morris PG (1981) The effects of NMR exposure on living organisms. I. A microbiological assay. Br J Radiol 54:615 34. Schwartz JL, Crooks LE (1982) NMR imaging produces no observable or cytotoxicity in mammalian cells. Am J Roentgenol 139:583 35. Wolff S, Crooks LE, Brown P (1980) Tests for DNA and chromosomal damage induced by nuclear magnetic resonance imaging. Radiology 136:707 36. Cook P, Morris PG (1981) The effects of NMR exposure on living organisms. II. A genetic study of human lymphocytes. Br J Radiol 54:622 37. Johnson IR, Symonds EM, Kean OM (1984) Imaging of the pregnant human uterus with nuclear magnetic resonance. Am J Obstet Gynecol 148:1136 38. Bydder GM, Steiner RE, Young IR (1982) Clinical NMR imaging of the brain: 140 cases. Am J Roentgenol 139: 215 39. Gramsen G, Webb WR, Sheldon P (1983) Nuclear magnetic resonance imaging of the thorax. Radiology 147:473 40. Kressel AY, Axel L, Thickman D (1983) NMR imaging of the abdomen at 0.12T: initial experience with a resistive magnet. Am J Roentgenol 141:1179 41. Buonocore E, Borkowski GP, Pavilecek W (1985) NMR imaging of the abdomen: technical considerations. Am J Roentgenol 141 : 1171 42. Bryan RT, Buller HE, LiPuma JP (1983) NMR scanning of the pelvis: initial experience with a 0.3T system. Am J Roentgenol 14l : 1111 43. Brady TJ, Rosen BR, Pykett IL (1983) NMR imaging of leg tumors. Radiology 149:181 44. Kerfkens RJ, Higgins CB, Hricak H (1983) Nuclear magnetic resonance imaging of the cardiovascular system: normal and pathological findings. Radiology 147:749 45. Weinreb JC, Low TW, Santos-Ramos R (1985) Magnetic resonance imaging in obstetric diagnosis. I. Radiology 154:157 46. McCarthy SM, Filly RA, Stark DO (1985) Obstetrical magnetic resonance imaging: fetal anatomy. I. Radiology 154:427 47. McCarthy SM, Filly RA, Stark DO (1985) Magnetic resonance imaging of fetal anomalies in utero: early experience. Am J Radiol 145:677 48. McCarthy SM, Filly RA, Stark DO (1985) Intrauterine growth retardation: evaluation by magnetic resonance (work in progress). Radiology 155:425 49. Honigan WC, Gibson J, Kleopouolos NJ (1986) Medical imaging of fetal ventriculomegaly. J Neurosurg 64:575 50. Smith FW (1985) The potential use of nuclear magnetic resonance imaging in pregnancy. J Perinat Med 13:265 51. Nuntz MC, Grossman RI, Isaacson G (1987) MR imaging of fetal brain. J Comput Assist Sonogr 11:120 52. McCarthy SM (1986) Magnetic resonance imaging in obstetrics and gynecology. Magn Reson Imgaging 4:59 53. Adzick NS, Harrison MR, Flake AW (1985) Development of a fetal renal function test using endogenous creatinine clearance. J Pediatr Surg 20:602 54. Kampmann J, Molhalen-Hansen J (1981) Glomerular ill-
55.
56.
57.
58.
59.
60.
61.
62.
63.
64.
65. 66. 67.
68.
69.
70. 71. 72. 73. 74. 75. 76. 77.
tration rate and creatinine clearance. Br J Clin Pharmacol 12:7-14 Arant BS, Edelman CM, Spitzer A (1972) The congruence of creatinine and inulin clearance in children: the use of the technicon auto analyzer. J Pediatr 81:559-561 Aperia A, Broberger O, Elinder G (1981) Postnatal development of renal function in pre-term and full-term infants. Acta Paediatr Scand 70: 183-187 Koszalka TR, Jensh RP, Brent RL (1975) Placental transport of creatinine in the rat. Proc Soc Exp BioI Med 148: 864 Hodari AA, Mariona FG, Houlihan RT (1973) Creatinine transport in the maternal-fetal complex. Obstet Gynecol 41 : 47 Pitkin RM, Reynolds WA (1975) Creatinine clearance between mother, fetus and amniotic fluid. Am J Physiol 228 : 231 McGaughey HS, Carey EL, Scoggin WA (1960) Creatinine transport between baby and mother at term. Am J Obstet Gynecol 80:108 Hutchinson DL, Bashore RA, Will DW (1962) Creatinine equilibrium between mother and nephrectomized primate fetus. Proc Soc Exp Biol Med 110:395 Shepard MJ, Richards VA, Berkowitz RL (1982) An evaluation of two equations for predicting fetal weight by ultrasound. Am J Obstet Gynecol 142:47 Robillard JE, Kulvinskas C, Sesscons C (1975) Maturational changes in fetal glomerular filtration rate; Am J Obstet Gynecol 122:601 Warshaw BL, Edelbrock HH, Ettinger RB (1982) Progression to end-stage renal disease in children with obstructive uropathy. J Pediatr 100: 183-187 McGrory WW (1972) Developmental nephrology. Harvard University Press, Cambridge, pp 40-46 Oliver J (1968) Nephrons and kidneys. Harper and Row, New York, pp 1-24 Bernstein J (1976) A classification of renal cysts. In: Gardner KD (ed) Cystic disease of the kidney. Wiley, New York, pp 7-30 Bernstein J (1971) The morphogenesis of renal parenchyreal maldevelopment (renal dysplasia). Pediatr Clin North Am 18:395 Felson B, Cussen LJ (1975) The hydronephrotic type of unilateral congenital multicystic disease of the kidney. Semin Roentgenol 10:113-123 Potter EL (1972) Normal and abnormal development of the kidney. Year Book, Chicago Harrison MR, Golbus MS, Filly RA (1987) Fetal hydronephrosis: selection and surgical repair. J Pediatr Surg 22: 556-558 McGrory WW (1972) Development of renal function in utero. Harvard University Press, Cambridge, pp 51-78 Alexander DP, Nixon DA (1961) The fetal kidney. Br Med Bull 17:112-117 McCance RA, Widdowson EM (1953) Renal function before birth. Proc R Soc Lond 141 : 448-497 Weinstein L, Anderson CF, Finley PR (1982) The in utero management of urinary outflow tract obstruction. J Clin Ultrasound 10:465-468 McFadyen IR, Wigglesworth JS, Dillon MJ (1983) Fetal urinary tract obstruction: is active intervention before delivery indicated? Br J Obstet Gynaecol 90:342-349 Adzick NS, Harrison MR, Glick PL, Flake AW (1985) Fetal urinary tract obstruction: experimental pathophysiolo gy. Semin Perinatol 9:79-89
Received February 28, 1988; Accepted March 11, 1988