PEDIATRIC UROLOGY
NATURAL HISTORY OF FETAL AND NEONATAL HYDRONEPHROSIS LOWELL R. KING, M.D. PAUL A. HATCHER, M.D. From the Section of Pediatric Urology, Division of Urology, Duke University Medical Center, Durham, North Carolina
ABSTRACT-Since pyelocalicectasis alone is common in fetuses, we reviewed reports of fetal hydronephrosis that resolved spontaneously or at birth. Severe fetal hydronephrosis with calicectasis or parenchymal thinning rarely resolves spontaneously before or after birth. We also reviewed the clinical and experimental literature on renal hypertrophy. After unilateral nephrectomy in neonatal animals or after birth with congenital absence of one kidney in humans, the remaining kidney hypertrophies very quickly. In infants and young animals, the eventual size of the remaining kidney is inversely proportional to the age at which one kidney is lost. This improvement in residual renal junction seen after renal loss in infancy, compared with older children, itself constitutes a strong argument for early relief of obstruction. If contralateral renal hypertrophy has occurred, the treated damaged kidney may resume growth in parallel with its hypertrophied mate but does not become as large or recover normal potential for growth. In other words, if correction of a unilateral obstruction is deferred until contralateral hypertrophy occurs, the obstructed kidney then has less potential for recovery of function.
Widespread screening fetal ultrasonography has had an immense impact on pediatric urology. Fetuses with persistent bladder distention generally prove to have urethral obstruction or agenesis, or the prune belly syndrome. l Ureteroceles may be diagnosed presumptively in the fetus. Such obstructions can be defined immediately after birth and be removed before infection occurs. Renal function may be compromised by associated renal dysplasia, but all agree that renal decompression at a few days of life (if not in utero) optimizes the potential for the development of renal function in such patients. 2 The optimal management of fetuses with hydroureteronephrosis or pyelocalicectasis Without bladder obstruction is not as clear. When unilateral, the condition is not lifethreatening, but proper treatment must be aimed at maximizing renal recovery. Disagreement over management occurs for two major reasons. Fetal "hydronephrosIs" often has been
lil\()LOGY
\IAY Hmo
/
VOLU\IE XXX\'. NUMBEH 5
reported to disappear or resolve spontaneously during later fetal life or found to be absent at birth. Postnatal evaluation may reveal that persisting dilatation is due to reflux or megacalycosis with or without nonobstructed megaureter, not to obstruction. Due to immaturity of the kidney at birth, imaging by intravenous urography is often not diagnostic even in the absence of obstruction so some form of renal scan, using furosemide (Lasix) to generate a renogram, or wash-out curve, is usually employed to differentiate between obstructed systems and simple dilatation. However, the method of performing a "furosemide renogram" has not been completely standardized, and sometimes varies within the same institution. Some have questioned the reliability of interpretation of such studies perhaps due to the "immaturity" of neonatal renal function. Given these concerns, is it not reasonable to do a cystogram to diagnose or exclude reflux and defer upper tract evaluation for a few weeks or
433
months until the normal kidney functions more efficiently? May fetal hydronephrosis still not resolve as it is seen to do so often in utero? We will address these questions by reviewing the current literature which shows that spontaneous resolution of severe hydronephrosis (with calicectasis or parenchymal thinning) is rare in utero and must be even less likely after birth. Additionally, clear experimental and clinical data exist which demonstrate that contralateral hypertrophy in the normal kidney opposite a possibly obstructed mate begins quickly after birth, at least when the hydronephrotic kidney has measurably reduced function, and that the potential for such hypertrophy is greatest at birth. A damaged kidney, if not dysgenebc or dysplastic, may resume growth if an obstructive problem can be corrected, but contralateral hypertrophy in the "normal" kidney persists. In other words, the hypertrophied kidney does not "give back" function to its impaired mate; the paired organs resume growth in parallel when a unilateral problem is corrected. The implication for a patient with unilateral renal obstruction is clear. Once contralateral hypertrophy has occurred, the potential for return of function in the obstructed organ is reduced. Since the degree of hypertrophy increases with the duration of the stimulus, the longer a unilateral obstruction persists, the less the potential for recovery in that organ. Every effort should, therefore, be made to diagnose unilateral obstruction precisely in the perinatal period since repair of the obstruction or decompression at that time will optimize the potential for ipsilateral renal growth. Enough experience has now been gained with pyeloplasty in very young, even premature, infants to be certain that the young age at the time of surgery does not prejudice the outcome. Indeed, surgical results may be better than when such surgery is deferred to six months or one year of age.
Calyceal dilation
6
Grade I
t
Grade II
t e
Grade I I I
Grade IV
Grade V
C
PhYSiological
S1 ze oc pelvis 1 cm
Normal calyces
1 - 1. 5 em
Sliqht dilation
> 1. 5 em
Moderate dilation
) 1. 5 em
Severe dilation + atrophic cortex
> 1. 5
em
FIGURE 1. Most useful system for grading fetal hydronephrosis. (From Grignon et al.,4 with permission.)
With the advent of fetal ultrasound bladder outlet obstruction, hydroureteronephrosis and pyelocalicectasis began to be detected in utero and are now found in 1 in 600 to 800 fetuses. Bladder obstruction is presumed when the fetal bladder is large and is seen not to empty on sequential studies. Hydronephrosis and oligohydramnios mayor may not be associated. It is in those with profound oligohydramnios that fetal decompression is sometimes deemed desirable,
since amniotic fluid is necessary for development of the lungs. 3 Most such infants are eval uated and treated a few days after birth. Hydroureteronephrosis may be due to reflux, obstructed megaureter, or nonobstructed megaureter, with or without megacalycosis (dilated calices). Dilatation without obstruction is now presumed to result from obstruction during fetal life which resolved spontaneously. Pyelocalicectasis is the most common anomaly of the urinary tract to be imaged on 'fetal ultrasound, and is the most common type of dilatation observed to resolve spontaneously either later in pregnancy or at birth. Two systems for grading fetal hydronephrosis have been introduced. That of Grignon et al. 4 (Fig. 1) encompasses those with detectable renal pelvic dilatation (Grade I), pelvic dilatation greater than 1 cm (Grade II), and three degrees exhibiting progressive calicectasis (Grades III - IV). Arger et al., 5 employ a similar system for Grades I and II. Their Grade III is fetuses with parenchymal cysts, who usually prove to have multicystic kidneys or at least small kidneys with persistent cystic dysplasia. Postnatal ultrasonography should be deferred in these patients until tWO days of age, as Laing et al. 6 have shown (by ultrasound) that fetal hydronephrosis may be absent at birth but return thereafter. 1 In the Grignon and in Arger reports all but. of 44 infants with Grade I fetal hydronephros1i were normal after birth. When the degree ~_ pelvic dilatation was greater (Grade II), the dl) atation was more likelv to persist (Table I.' When calicectasis was present, hydronephrosIS
434
UROLOGY
Fetal Hydronephrosis
/
MAY J 9DO
VOLUME
xxxv.
NUMBEH 5
TABLE
I.
Resolving fetal hydronephrosis
Degree of Hydronephrosis Moderate Grignon et al. 4 (1986) Sanders and Graham 8 (1982) Arger et al. 5 (1985) Severe Grignon et al. 4 (1986) Baker et al. 9 (1985) Grignon et ai. 7 (1986) Arger et az.s (1985) Unspecified Kleiner et al. lO (1987) Blane et al. ll (1983) Hobbins et al. 12 (1984) Hellstrom et al. 13 (1984) Nicolini et al. 14 (1987) Harrison et al. 15 (1982) Smith et al. 16 (1987) Wilkins et alY (1987)
No. of Pts.
No. Resolved
31 12 8
15 8 1
32 1 39
2 1 1
9
o
25 29 25 13 30 26
o
20 9
1 3 5
13 3 9 1
rarely resolved (Table I). In the Grignon et al. 7 series, for instance, all with Grades IV and V pyelocalicectasis had evidence of obstruction on furosemide renogram and eventually required surgical correction. Unfortunately, in most reports of resolving fetal hydronephrosis the presence or absence of calicectasis or cortical thinning is not reported (Table I), and it is impossible to determine the degree of dilatation that was present except in an occasional case which is illustrated. However, it seems clear that more than moderate calicectasis seldom resolves even during intrauterine life. At this time, we know of no infant found to have obstruction after birth on the basis of a diuretic renogram in whom the obstruction later resolved. Further, if calicectasis is present at birth in an unequivocally unobstructed system, .such calicectasis (megacalycosis) is very apt to persist, at least in some degree. Importance of Contralateral Renal Hypertrophy When the accuracy of the diuretic renogram comes into question, the diagnostic dilemma may be resolved by a Whitaker test, or a related pressure perfusion study in neonates as in older children. Such tests are invasive, require a general anesthetic, and usually will not be recommended unless the urologist believes it is important to correct obstruction soon after birth. The rapid onset of renal hypertrophy, and the
UROLOGY
~1AY
1990
i
VOLUME XXXV NUMBER 5
effect of contralateral hypertrophy on the potential for recovery of an obstructed kidney comprise compelling arguments in this regard. It has been known since antiquity that animals and humans can survive and lead a normal life with only one kidney. In the human, nephrogenesis is complete before term.n Ogden in 1967,18 following 28 adult renal transplant donors two to four years after nephrectomy, found that renal function had returned to approximatley 70 percent of normal. MacKay, MacKay, and Addis 19 had already shown that renal hyperplasia and hypertrophy were the phenomenon which produced the increased function in the remaining kidney. The hypertrophy occurs rapidly after unilateral nephrectomy, and is virtually complete a year later. Kaufman et al. 20 noted the remarkable degree of compensatory change in renal mass and function following the removal of more than one kidney, and demonstrated that the extent of compensatory renal adaptation correlates directly with the amount of renal tissue removed. This group21 later (1975) reported that in Sprague-Dawley rats the mean nephron GFR increased 60 percent after unilateral nephrectomy and 150 percent after ablation of 75 percent of renal tissue. These changes paralleled increases in renal growth (hypertrophy) under the same conditions. Mean glomerular blood flow rose 90 percent and 240 percent, respectively, resulting in a progressive fall in the filtration fraction and setting in motion a chain of internal mechanisms used by the hypertrophied kidney to maintain homeostasis. Galla, Klein-Robbenhaar, and Hayslett 22 nephrectomized weanling rats (50-80 g) and young adults. One month after nephrectomy the percent change in remaining renal mass was 144 percent in the weanling animals and 66 percent in the adults. Larsson, Aperia, and Wilton 23 extended these observations, nephrectomizing animals at five, twelve, and forty days of age. Kidney weight, GFR, single nephron GFR, and other parameters were assessed at sixty days of age, Hayslett having shown that in these rats the weight of the hypertrophied kidney reaches a plateau fourteen days after nephrectomy. The GFR increased in proportion to the size of the remaining kidney. Age-dependent potentiation of compensatory increase in the size of the remaining kidney and GFR was much more pronounced when nephrectomy was performed at five days of age. Eventual kidney weight was 21 percent greater and
435
GFR 23 percent higher in rats nephrectomized at five days compared with twelve days of age. Those nephrectomized at twelve days still exhibited 8 percent better renal weight and functional parameters than the animals operated on at forty days, making the difference between the infant and weanling even more pronounced. Aschinberg et al. 24 studied puppies that underwent removal of 75 percent of renal mass before one week of age or at eight weeks of age. Sham-operated littermates served as controls. These controls, examined six weeks after birth, experienced a 2.5-fold increase in kidney weight as a result of normal growth, whereas the group less than one week old at operation exhibited a 7.5-fold increase in the weight of the remaining tissue. By comparison, animals operated on at eight weeks experienced only a 3.5-fold increase in the weight of their remaining kidney, while sham-operated littermates doubled their renal mass during the same period of observation. Therefore, compensatory growth accounted for a five-fold increase in the nephrectomized newborns against only a 1.5fold increase in the juvenile animals. Animals that underwent a three-quarter reduction in renal mass at birth reached a level of GFR that was not significantly different from shamoperated littermates by six weeks of age. In contrast, the dogs operated on at eight weeks had a GFR six weeks later which was only 45 percent of that observed in the controls. In an animal model more analogous to human partial obstruction, Claesson 25 (1987) produced partial ureteral obstruction in one- to three-day-old Sprague-Dawley rats by the method of Ulm and Miller, burying the ureter in the psoas muscle. Renal function was subsequently measured by carefully standardized renal scintigraphy using 99mTc-dimercaptosuccinic acid (DMSA) as well as renal weight and more conventional function tests. This monograph deserves review, since many aspects of hydronephrosis were studied and the bibliography is extensive. Among other things, Claesson found that release of obstruction after only two days resulted, three weeks later, in a slight reduction in ipsilateral renal weight which normalized after six weeks. However, from the third week there was a weight increase in the contralateral side which was still present at six weeks, after the side operated on had become normal. Release of the obstruction after seven days did not prevent a decrease in the weight of 436
the parenchyma on the obstructed side and a compensatory weight increase on the contralateral side. Again, changes in individual renal function paralleled changes in individual parenchymal weight. In babies born with a m ulticystic kidney, renal measurements of a normal contralateral kidney are the same as in neonates with two kidneys.26 There is no stimulus to hypertrophy before birth presumably because the placenta is excreting fetal metabolic wastes. Hypertrophy-reflected as accelerated renal growth-can be detected sonographically in a normal solitary kidney by nine days of age. 27 Judged by change in renal size, hypertrophy may continue until two to three years of age, but is 90 percent complete by four months. Aperia et al. 28 studied renal growth in the remaining kidney after removal of a contralateral kidney in childhood. More than three years after nephrectomy, the remaining kidneys were 35 percent to 65 percent larger than normal. The growth achieved by the remaining kidney was inversely related to the age at the time of nephrectomy. In other words, the younger the patient at time of nephrectomy, the more the remaining normal kidney was able to grow. Claesson et al. in 198Pg followed up 26 children with unilateral renal scarring from reflux nephropathy. These investigators used renal planimetry to measure renal area.' This method correlates better with renal mass than simpler estimates of renal length, especially in scarred kidneys. The total renal area (both kidneys) remained 97-98 percent of normal because of contralateral renal hypertrophy. With prevention of infection and/or correction of reflux both kidneys grew, but renal mass became normal in only two scarred kidneys. Without infections the kidneys grew in tandem, but the scarred kidney did not recover the function which had been compensated for by contralateral hypertrophy. The hypertrophied kidney remained larger than its scarred mate years after bilateral renal growth had resumed. Similar observations were made by Wilton et aI., 31 following up children with unilateral reflux nephropathy or after unilateral heminephrectomy. They also found that the smaller kidneys could resume growth, but remained smaller than a hypertrophied mate. These are key observations, since they prove that once contralateral hypertrophy has occurred the unilaterally affected kidney, damaged by obstruction, reflux, or infection cannot
UROLOGY
/
""lAY U)90
I
VOLUME XXXV, ""CMBER 5
TABLE
II.
Results oj pyeloplasty in early injancy « 3 months oj age) Ko. of Pyeloplasties
Series
Kin g
32 (1988) Flake et aZ. 33 (1986) Thorup et aZ. 34 (1985)
24 22 9
Hanna and Gluck 35 (1988) Wolpert et a1.'6 (1988) Kayle and Ehrlich 37 (1988)
25 34 20
Results No reoperation All successful 2 reoperations, 1 worse All improved No reoperation All improved or stable
become normal even if the disease process is completely corrected. The onset of contralateral hypertrophy reduces the potential for recovery in the affected kidney, and hypertrophy occurs very quickly after birth. Additionally, such hypertrophy is most extreme in young infants. Together, these studies of renal hypertrophy provide a compelling reason for early post-natal diagnosis and correction of unilateral obstruction. If a good result is achieved, as it usually is, the potential for return in function in the obstructed kidney is optimized (Table II). Summary It has been shown that there is great potential for return of function when unilateral obstruction is relieved in neonates and very young infants.38 There are two reasons for this. The nondysplastic neonatal kidney has a much greater potential capacity for hypertrophy than is present in older infants, children, or adults. Solitary kidneys are normal in size at birth since hypertrophy does not occur in utero. However, if one kidney only is abnormal at birth, the contralateral kidney begins to hypertrophy very quickly. Once this has occurred and, for instance, if reflux is corrected or UTI prevented, both kidneys will begin to grow, but the affected kidney remains smaller than its initially normal but subsequently hypertrophied mate. In other words, once unilateral hypertrophy has occurred it does not regress. This means that unilateral renal obstruction should be corrected as soon as it can be diagnosed in neonates and very young infants. Otherwise contralateral hypertrophy will occur and reduce the potential for recovery of function in the obstructed kidney. Durham, North Carolina 27710 (DR. KING)
UROLOGY
I\IAY 1990
VOLC'v1E XXXv l\CI\1BEH .5
References 1. International Fetal SllTgery Hegistry: Catheter shunts for fetal hydronephrosis and hydrocephalus, ;\; Engl J Mcd 315: 336 (1986). 2. Chevalier RL, and EI Dahr S: The cause for early relief of obstruction in young infants. in King LR (Ed): crologic Surgery in Neonates and Young Infants, Philadelphia, 'vB Saunders, chap 6, 1988, pp 95-118. 3. Glick DL, et al: Management of the fetus with congenital hydronephrosis, II. prognostic criteria and selection for treatment, J Pediatr Surg 20: 276 (198,5). 4. Grignon A, et al: Urinary tract dilatation in utero: classification and clinical applications. Radiology 160: 645 (1986). 5. Arger PH, et al: Routine fetal genitourinary tract screening, Radiology 1.56: 485 (1985). 6. Laing FC, et al: Postpartum evaluation of fetal hydronephrosis: optimal timing for followup sonography, Radiology 152: 423 (1984). 7. Grignon A, et ale Ureteropelvic junction stenosis: antenatal ultrasonographic diagnosis, postnatal investigation and followup, Radiology 160: 649 (1986). 8. Sanders R, and Graham D: Twelve cases of hydronephrosis in utero diagnosed by ultrasonography, J Ultrasound Med 1: 341 (1982). 9. Baker 'viE, Rosenberg ER, Bowie JD, and Gall S: Transient in utero hydronephrosis, J Ultrasound Med 4: .51 (1985). 10. Kleiner B, Callen PW, and Filly RW: Sonographic analysis of the fetus with ureteropelvic junction obstruction, Am J Radiol 148: 359 (1987). 11. Blane CE, Koff SA, Bowerman RA, and Barr M: Nonobstructive fetal hydronephrosis: sonographic recognition and therapeutic implications, Radiology 147: 95 (1983). 12. Hobbins JC, et al: Antenatal diagnosis of renal anomalies with ultrasound, I. obstructive uropathy, Am J Obstet Gynecol 148: 868 (1984). 13. Hellstrom WI. et ale The natural history of prenatal hydronephrosis with normal amounts of amniotic fluid, J Urol132: 947 (1984). 14. Nicolini V, et al: Perinatal management of fetal hydronephosis with normal bladder, J Perinatal Med 15: 54 (1987). 15. Harrison MR, et al: Fetal surgery for congenital hydronephrosis, Med Intell 306: 591 (1982). 16. Smith 0, Eggington JA, and Brookfield OS: Detection of abnormalities of fetal urinary tract as a predictor of renal tract disease, Br 'vied J 294: 27 (1987). 17. Wilkins lA, et al: The nonpredictive value of fetal urinary electrolytes: preliminary report of outcome and correlation with pathologic diagnosis, Am J Obstet Gynecol157: 694 (1987). 18. Ogden DA: Donor and recipient function 2 to 4 years after renal homotransplantation, Ann Intern Med 67: 998 (1967). 19. MacKay EM, MacKay LL, and Addis T: The degree of compensatory renal hypertrophy following unilateral nephrectomy, J Exp Med 56: 25.5 (1932). 20. Kaufman JM, et al: Compensatory adaptation of structure and function following progressive renal ablation, Kidney Int 6: 10 (1974). 21. Kaufman JM, Siegel NJ, and Hayslett JP: Functional and hemodynamic adaptation to progressive renal ablation, Circulation Res 36: 286 (1975). 22. Galla JH, Klein-Robbenhaar T, and Hayslett JP: Influence of age on the compensatory response in growth and function to unilateral nephrectomy, Yale J BioI Med 47: 218 (1974). 23. Larsson L, Aperia A, and Wilton P: Effect of normal development on compensatory renal growth, Kidney Int 18: 29 (1980) . 24. Aschinberg LC, et al: The influence of age on the response to renal parenchymal loss, Yale J BioI Med 51: 341 (1978). 25. Claesson G: Experimental hydronephrosis in newborn rats: effects on renal morphology and function, Kongl Carolinska Medico Chir Inst Stockholm, 1987. 26. Laufer 1. and Griscom :--JT: Compensatory renal hvpertrophy: absence in utero and development in early life, AJR 113: 464 (1971).
437
27. Chevalier RL, Campbell F, and Brenbridge ANAG: Nephrosonography and renal scintigraphy in evaluation of newborn with renomegaly, Urology 24: 96 (1984). 28. Aperia A, et aI: Renal growth and function in patients nephrectomized in childhood, Acta Pediatr Scand 66: 185 (1977). 29. Claesson I, et aI: Compensatory kidney growth in children with urinary tract infection and unilateral renal scarring: an epidemiologic study, Kidney Int 20: 759 (1981). 30. Claesson I, and Jacobsson B: Standardisierte messungen der nierenparenchymdicke. Methode und diagnosticher wert bei kindem, in Olbing H (Ed): Rezidivierende nicht obstruktive harnwegserkrankungen bei kindem, Heidelberg, SpringerVerlag, 1980, pp 65-75. 31. Wilton P, et aI: Renal compensatory hypertrophy in children with unilateral renal disease, Acta Paediatr 69: 83 (1980). 32. King LR, The management of multicystic kidney and ureteropelvic junction obstruction, in King LR (Ed): Urologic Sur-
gery in Neonates and Young Infants, Philadelphia, WB Saunders, chap 9, 1988, pp 140-154. 33. Flake AW, et aI: Ureteropelvic junction obstruction in the fetus, J Pediatr Surg 21: 1058 (1986). 34. Thorup J, et al: The prognosis of surgically treated congenital hydronephrosis after diagnosis in utero, J Urol 134: 914 (1985). 35. Hanna MK, and Gluck R: Ureteropelvic junction obstruction during the first year of life, Urology 31: 41 (1988). 36. Wolpert JJ, Woodward JR, and Parrott TS: Pyeloplasty in the young infant, presented at Section on Pediatric Urology, American Academy of Pediatrics, San Francisco, October 16, 1988. 37. Koyle MA, and Ehrlich RM: Management of ureteropelvic junction obstruction in neonate, Urology 31: 496 (1988). 38. King LR, et aI: The case for immediate pyeloplasty in the neonate with upJ obstruction, J Uro1132: 725 (1984).
438
UROLOGY
I
MAY 1990
VOLUME XXXV, NUMBEH 5
NATURAL HISTORY OF FETAL AND NEONATAL HYDRONEPHROSIS LOWELL R. KING, M.D. PAUL A. HATCHER, M.D. From the Section of Pediatric Urology, Division of Urology, Duke University Medical Center, Durham, North Carolina
ABSTRACT-Since pyelocalicectasis alone is common in fetuses, we reviewed reports of fetal hydronephrosis that resolved spontaneously or at birth. Severe fetal hydronephrosis with calicectasis or parenchymal thinning rarely resolves spontaneously before or after birth. We also reviewed the clinical and experimental literature on renal hypertrophy. After unilateral nephrectomy in neonatal animals or after birth with congenital absence of one kidney in humans, the remaining kidney hypertrophies very quickly. In infants and young animals, the eventual size of the remaining kidney is inversely proportional to the age at which one kidney is lost. This improvement in residual renal junction seen after renal loss in infancy, compared with older children, itself constitutes a strong argument for early relief of obstruction. If contralateral renal hypertrophy has occurred, the treated damaged kidney may resume growth in parallel with its hypertrophied mate but does not become as large or recover normal potential for growth. In other words, if correction of a unilateral obstruction is deferred until contralateral hypertrophy occurs, the obstructed kidney then has less potential for recovery of function.
Widespread screening fetal ultrasonography has had an immense impact on pediatric urology. Fetuses with persistent bladder distention generally prove to have urethral obstruction or agenesis, or the prune belly syndrome. l Ureteroceles may be diagnosed presumptively in the fetus. Such obstructions can be defined immediately after birth and be removed before infection occurs. Renal function may be compromised by associated renal dysplasia, but all agree that renal decompression at a few days of life (if not in utero) optimizes the potential for the development of renal function in such patients. 2 The optimal management of fetuses with hydroureteronephrosis or pyelocalicectasis Without bladder obstruction is not as clear. When unilateral, the condition is not lifethreatening, but proper treatment must be aimed at maximizing renal recovery. Disagreement over management occurs for two major reasons. Fetal "hydronephrosIs" often has been
lil\()LOGY
\IAY Hmo
/
VOLU\IE XXX\'. NUMBEH 5
reported to disappear or resolve spontaneously during later fetal life or found to be absent at birth. Postnatal evaluation may reveal that persisting dilatation is due to reflux or megacalycosis with or without nonobstructed megaureter, not to obstruction. Due to immaturity of the kidney at birth, imaging by intravenous urography is often not diagnostic even in the absence of obstruction so some form of renal scan, using furosemide (Lasix) to generate a renogram, or wash-out curve, is usually employed to differentiate between obstructed systems and simple dilatation. However, the method of performing a "furosemide renogram" has not been completely standardized, and sometimes varies within the same institution. Some have questioned the reliability of interpretation of such studies perhaps due to the "immaturity" of neonatal renal function. Given these concerns, is it not reasonable to do a cystogram to diagnose or exclude reflux and defer upper tract evaluation for a few weeks or
433
months until the normal kidney functions more efficiently? May fetal hydronephrosis still not resolve as it is seen to do so often in utero? We will address these questions by reviewing the current literature which shows that spontaneous resolution of severe hydronephrosis (with calicectasis or parenchymal thinning) is rare in utero and must be even less likely after birth. Additionally, clear experimental and clinical data exist which demonstrate that contralateral hypertrophy in the normal kidney opposite a possibly obstructed mate begins quickly after birth, at least when the hydronephrotic kidney has measurably reduced function, and that the potential for such hypertrophy is greatest at birth. A damaged kidney, if not dysgenebc or dysplastic, may resume growth if an obstructive problem can be corrected, but contralateral hypertrophy in the "normal" kidney persists. In other words, the hypertrophied kidney does not "give back" function to its impaired mate; the paired organs resume growth in parallel when a unilateral problem is corrected. The implication for a patient with unilateral renal obstruction is clear. Once contralateral hypertrophy has occurred, the potential for return of function in the obstructed organ is reduced. Since the degree of hypertrophy increases with the duration of the stimulus, the longer a unilateral obstruction persists, the less the potential for recovery in that organ. Every effort should, therefore, be made to diagnose unilateral obstruction precisely in the perinatal period since repair of the obstruction or decompression at that time will optimize the potential for ipsilateral renal growth. Enough experience has now been gained with pyeloplasty in very young, even premature, infants to be certain that the young age at the time of surgery does not prejudice the outcome. Indeed, surgical results may be better than when such surgery is deferred to six months or one year of age.
Calyceal dilation
6
Grade I
t
Grade II
t e
Grade I I I
Grade IV
Grade V
C
PhYSiological
S1 ze oc pelvis 1 cm
Normal calyces
1 - 1. 5 em
Sliqht dilation
> 1. 5 em
Moderate dilation
) 1. 5 em
Severe dilation + atrophic cortex
> 1. 5
em
FIGURE 1. Most useful system for grading fetal hydronephrosis. (From Grignon et al.,4 with permission.)
With the advent of fetal ultrasound bladder outlet obstruction, hydroureteronephrosis and pyelocalicectasis began to be detected in utero and are now found in 1 in 600 to 800 fetuses. Bladder obstruction is presumed when the fetal bladder is large and is seen not to empty on sequential studies. Hydronephrosis and oligohydramnios mayor may not be associated. It is in those with profound oligohydramnios that fetal decompression is sometimes deemed desirable,
since amniotic fluid is necessary for development of the lungs. 3 Most such infants are eval uated and treated a few days after birth. Hydroureteronephrosis may be due to reflux, obstructed megaureter, or nonobstructed megaureter, with or without megacalycosis (dilated calices). Dilatation without obstruction is now presumed to result from obstruction during fetal life which resolved spontaneously. Pyelocalicectasis is the most common anomaly of the urinary tract to be imaged on 'fetal ultrasound, and is the most common type of dilatation observed to resolve spontaneously either later in pregnancy or at birth. Two systems for grading fetal hydronephrosis have been introduced. That of Grignon et al. 4 (Fig. 1) encompasses those with detectable renal pelvic dilatation (Grade I), pelvic dilatation greater than 1 cm (Grade II), and three degrees exhibiting progressive calicectasis (Grades III - IV). Arger et al., 5 employ a similar system for Grades I and II. Their Grade III is fetuses with parenchymal cysts, who usually prove to have multicystic kidneys or at least small kidneys with persistent cystic dysplasia. Postnatal ultrasonography should be deferred in these patients until tWO days of age, as Laing et al. 6 have shown (by ultrasound) that fetal hydronephrosis may be absent at birth but return thereafter. 1 In the Grignon and in Arger reports all but. of 44 infants with Grade I fetal hydronephros1i were normal after birth. When the degree ~_ pelvic dilatation was greater (Grade II), the dl) atation was more likelv to persist (Table I.' When calicectasis was present, hydronephrosIS
434
UROLOGY
Fetal Hydronephrosis
/
MAY J 9DO
VOLUME
xxxv.
NUMBEH 5
TABLE
I.
Resolving fetal hydronephrosis
Degree of Hydronephrosis Moderate Grignon et al. 4 (1986) Sanders and Graham 8 (1982) Arger et al. 5 (1985) Severe Grignon et al. 4 (1986) Baker et al. 9 (1985) Grignon et ai. 7 (1986) Arger et az.s (1985) Unspecified Kleiner et al. lO (1987) Blane et al. ll (1983) Hobbins et al. 12 (1984) Hellstrom et al. 13 (1984) Nicolini et al. 14 (1987) Harrison et al. 15 (1982) Smith et al. 16 (1987) Wilkins et alY (1987)
No. of Pts.
No. Resolved
31 12 8
15 8 1
32 1 39
2 1 1
9
o
25 29 25 13 30 26
o
20 9
1 3 5
13 3 9 1
rarely resolved (Table I). In the Grignon et al. 7 series, for instance, all with Grades IV and V pyelocalicectasis had evidence of obstruction on furosemide renogram and eventually required surgical correction. Unfortunately, in most reports of resolving fetal hydronephrosis the presence or absence of calicectasis or cortical thinning is not reported (Table I), and it is impossible to determine the degree of dilatation that was present except in an occasional case which is illustrated. However, it seems clear that more than moderate calicectasis seldom resolves even during intrauterine life. At this time, we know of no infant found to have obstruction after birth on the basis of a diuretic renogram in whom the obstruction later resolved. Further, if calicectasis is present at birth in an unequivocally unobstructed system, .such calicectasis (megacalycosis) is very apt to persist, at least in some degree. Importance of Contralateral Renal Hypertrophy When the accuracy of the diuretic renogram comes into question, the diagnostic dilemma may be resolved by a Whitaker test, or a related pressure perfusion study in neonates as in older children. Such tests are invasive, require a general anesthetic, and usually will not be recommended unless the urologist believes it is important to correct obstruction soon after birth. The rapid onset of renal hypertrophy, and the
UROLOGY
~1AY
1990
i
VOLUME XXXV NUMBER 5
effect of contralateral hypertrophy on the potential for recovery of an obstructed kidney comprise compelling arguments in this regard. It has been known since antiquity that animals and humans can survive and lead a normal life with only one kidney. In the human, nephrogenesis is complete before term.n Ogden in 1967,18 following 28 adult renal transplant donors two to four years after nephrectomy, found that renal function had returned to approximatley 70 percent of normal. MacKay, MacKay, and Addis 19 had already shown that renal hyperplasia and hypertrophy were the phenomenon which produced the increased function in the remaining kidney. The hypertrophy occurs rapidly after unilateral nephrectomy, and is virtually complete a year later. Kaufman et al. 20 noted the remarkable degree of compensatory change in renal mass and function following the removal of more than one kidney, and demonstrated that the extent of compensatory renal adaptation correlates directly with the amount of renal tissue removed. This group21 later (1975) reported that in Sprague-Dawley rats the mean nephron GFR increased 60 percent after unilateral nephrectomy and 150 percent after ablation of 75 percent of renal tissue. These changes paralleled increases in renal growth (hypertrophy) under the same conditions. Mean glomerular blood flow rose 90 percent and 240 percent, respectively, resulting in a progressive fall in the filtration fraction and setting in motion a chain of internal mechanisms used by the hypertrophied kidney to maintain homeostasis. Galla, Klein-Robbenhaar, and Hayslett 22 nephrectomized weanling rats (50-80 g) and young adults. One month after nephrectomy the percent change in remaining renal mass was 144 percent in the weanling animals and 66 percent in the adults. Larsson, Aperia, and Wilton 23 extended these observations, nephrectomizing animals at five, twelve, and forty days of age. Kidney weight, GFR, single nephron GFR, and other parameters were assessed at sixty days of age, Hayslett having shown that in these rats the weight of the hypertrophied kidney reaches a plateau fourteen days after nephrectomy. The GFR increased in proportion to the size of the remaining kidney. Age-dependent potentiation of compensatory increase in the size of the remaining kidney and GFR was much more pronounced when nephrectomy was performed at five days of age. Eventual kidney weight was 21 percent greater and
435
GFR 23 percent higher in rats nephrectomized at five days compared with twelve days of age. Those nephrectomized at twelve days still exhibited 8 percent better renal weight and functional parameters than the animals operated on at forty days, making the difference between the infant and weanling even more pronounced. Aschinberg et al. 24 studied puppies that underwent removal of 75 percent of renal mass before one week of age or at eight weeks of age. Sham-operated littermates served as controls. These controls, examined six weeks after birth, experienced a 2.5-fold increase in kidney weight as a result of normal growth, whereas the group less than one week old at operation exhibited a 7.5-fold increase in the weight of the remaining tissue. By comparison, animals operated on at eight weeks experienced only a 3.5-fold increase in the weight of their remaining kidney, while sham-operated littermates doubled their renal mass during the same period of observation. Therefore, compensatory growth accounted for a five-fold increase in the nephrectomized newborns against only a 1.5fold increase in the juvenile animals. Animals that underwent a three-quarter reduction in renal mass at birth reached a level of GFR that was not significantly different from shamoperated littermates by six weeks of age. In contrast, the dogs operated on at eight weeks had a GFR six weeks later which was only 45 percent of that observed in the controls. In an animal model more analogous to human partial obstruction, Claesson 25 (1987) produced partial ureteral obstruction in one- to three-day-old Sprague-Dawley rats by the method of Ulm and Miller, burying the ureter in the psoas muscle. Renal function was subsequently measured by carefully standardized renal scintigraphy using 99mTc-dimercaptosuccinic acid (DMSA) as well as renal weight and more conventional function tests. This monograph deserves review, since many aspects of hydronephrosis were studied and the bibliography is extensive. Among other things, Claesson found that release of obstruction after only two days resulted, three weeks later, in a slight reduction in ipsilateral renal weight which normalized after six weeks. However, from the third week there was a weight increase in the contralateral side which was still present at six weeks, after the side operated on had become normal. Release of the obstruction after seven days did not prevent a decrease in the weight of 436
the parenchyma on the obstructed side and a compensatory weight increase on the contralateral side. Again, changes in individual renal function paralleled changes in individual parenchymal weight. In babies born with a m ulticystic kidney, renal measurements of a normal contralateral kidney are the same as in neonates with two kidneys.26 There is no stimulus to hypertrophy before birth presumably because the placenta is excreting fetal metabolic wastes. Hypertrophy-reflected as accelerated renal growth-can be detected sonographically in a normal solitary kidney by nine days of age. 27 Judged by change in renal size, hypertrophy may continue until two to three years of age, but is 90 percent complete by four months. Aperia et al. 28 studied renal growth in the remaining kidney after removal of a contralateral kidney in childhood. More than three years after nephrectomy, the remaining kidneys were 35 percent to 65 percent larger than normal. The growth achieved by the remaining kidney was inversely related to the age at the time of nephrectomy. In other words, the younger the patient at time of nephrectomy, the more the remaining normal kidney was able to grow. Claesson et al. in 198Pg followed up 26 children with unilateral renal scarring from reflux nephropathy. These investigators used renal planimetry to measure renal area.' This method correlates better with renal mass than simpler estimates of renal length, especially in scarred kidneys. The total renal area (both kidneys) remained 97-98 percent of normal because of contralateral renal hypertrophy. With prevention of infection and/or correction of reflux both kidneys grew, but renal mass became normal in only two scarred kidneys. Without infections the kidneys grew in tandem, but the scarred kidney did not recover the function which had been compensated for by contralateral hypertrophy. The hypertrophied kidney remained larger than its scarred mate years after bilateral renal growth had resumed. Similar observations were made by Wilton et aI., 31 following up children with unilateral reflux nephropathy or after unilateral heminephrectomy. They also found that the smaller kidneys could resume growth, but remained smaller than a hypertrophied mate. These are key observations, since they prove that once contralateral hypertrophy has occurred the unilaterally affected kidney, damaged by obstruction, reflux, or infection cannot
UROLOGY
/
""lAY U)90
I
VOLUME XXXV, ""CMBER 5
TABLE
II.
Results oj pyeloplasty in early injancy « 3 months oj age) Ko. of Pyeloplasties
Series
Kin g
32 (1988) Flake et aZ. 33 (1986) Thorup et aZ. 34 (1985)
24 22 9
Hanna and Gluck 35 (1988) Wolpert et a1.'6 (1988) Kayle and Ehrlich 37 (1988)
25 34 20
Results No reoperation All successful 2 reoperations, 1 worse All improved No reoperation All improved or stable
become normal even if the disease process is completely corrected. The onset of contralateral hypertrophy reduces the potential for recovery in the affected kidney, and hypertrophy occurs very quickly after birth. Additionally, such hypertrophy is most extreme in young infants. Together, these studies of renal hypertrophy provide a compelling reason for early post-natal diagnosis and correction of unilateral obstruction. If a good result is achieved, as it usually is, the potential for return in function in the obstructed kidney is optimized (Table II). Summary It has been shown that there is great potential for return of function when unilateral obstruction is relieved in neonates and very young infants.38 There are two reasons for this. The nondysplastic neonatal kidney has a much greater potential capacity for hypertrophy than is present in older infants, children, or adults. Solitary kidneys are normal in size at birth since hypertrophy does not occur in utero. However, if one kidney only is abnormal at birth, the contralateral kidney begins to hypertrophy very quickly. Once this has occurred and, for instance, if reflux is corrected or UTI prevented, both kidneys will begin to grow, but the affected kidney remains smaller than its initially normal but subsequently hypertrophied mate. In other words, once unilateral hypertrophy has occurred it does not regress. This means that unilateral renal obstruction should be corrected as soon as it can be diagnosed in neonates and very young infants. Otherwise contralateral hypertrophy will occur and reduce the potential for recovery of function in the obstructed kidney. Durham, North Carolina 27710 (DR. KING)
UROLOGY
I\IAY 1990
VOLC'v1E XXXv l\CI\1BEH .5
References 1. International Fetal SllTgery Hegistry: Catheter shunts for fetal hydronephrosis and hydrocephalus, ;\; Engl J Mcd 315: 336 (1986). 2. Chevalier RL, and EI Dahr S: The cause for early relief of obstruction in young infants. in King LR (Ed): crologic Surgery in Neonates and Young Infants, Philadelphia, 'vB Saunders, chap 6, 1988, pp 95-118. 3. Glick DL, et al: Management of the fetus with congenital hydronephrosis, II. prognostic criteria and selection for treatment, J Pediatr Surg 20: 276 (198,5). 4. Grignon A, et al: Urinary tract dilatation in utero: classification and clinical applications. Radiology 160: 645 (1986). 5. Arger PH, et al: Routine fetal genitourinary tract screening, Radiology 1.56: 485 (1985). 6. Laing FC, et al: Postpartum evaluation of fetal hydronephrosis: optimal timing for followup sonography, Radiology 152: 423 (1984). 7. Grignon A, et ale Ureteropelvic junction stenosis: antenatal ultrasonographic diagnosis, postnatal investigation and followup, Radiology 160: 649 (1986). 8. Sanders R, and Graham D: Twelve cases of hydronephrosis in utero diagnosed by ultrasonography, J Ultrasound Med 1: 341 (1982). 9. Baker 'viE, Rosenberg ER, Bowie JD, and Gall S: Transient in utero hydronephrosis, J Ultrasound Med 4: .51 (1985). 10. Kleiner B, Callen PW, and Filly RW: Sonographic analysis of the fetus with ureteropelvic junction obstruction, Am J Radiol 148: 359 (1987). 11. Blane CE, Koff SA, Bowerman RA, and Barr M: Nonobstructive fetal hydronephrosis: sonographic recognition and therapeutic implications, Radiology 147: 95 (1983). 12. Hobbins JC, et al: Antenatal diagnosis of renal anomalies with ultrasound, I. obstructive uropathy, Am J Obstet Gynecol 148: 868 (1984). 13. Hellstrom WI. et ale The natural history of prenatal hydronephrosis with normal amounts of amniotic fluid, J Urol132: 947 (1984). 14. Nicolini V, et al: Perinatal management of fetal hydronephosis with normal bladder, J Perinatal Med 15: 54 (1987). 15. Harrison MR, et al: Fetal surgery for congenital hydronephrosis, Med Intell 306: 591 (1982). 16. Smith 0, Eggington JA, and Brookfield OS: Detection of abnormalities of fetal urinary tract as a predictor of renal tract disease, Br 'vied J 294: 27 (1987). 17. Wilkins lA, et al: The nonpredictive value of fetal urinary electrolytes: preliminary report of outcome and correlation with pathologic diagnosis, Am J Obstet Gynecol157: 694 (1987). 18. Ogden DA: Donor and recipient function 2 to 4 years after renal homotransplantation, Ann Intern Med 67: 998 (1967). 19. MacKay EM, MacKay LL, and Addis T: The degree of compensatory renal hypertrophy following unilateral nephrectomy, J Exp Med 56: 25.5 (1932). 20. Kaufman JM, et al: Compensatory adaptation of structure and function following progressive renal ablation, Kidney Int 6: 10 (1974). 21. Kaufman JM, Siegel NJ, and Hayslett JP: Functional and hemodynamic adaptation to progressive renal ablation, Circulation Res 36: 286 (1975). 22. Galla JH, Klein-Robbenhaar T, and Hayslett JP: Influence of age on the compensatory response in growth and function to unilateral nephrectomy, Yale J BioI Med 47: 218 (1974). 23. Larsson L, Aperia A, and Wilton P: Effect of normal development on compensatory renal growth, Kidney Int 18: 29 (1980) . 24. Aschinberg LC, et al: The influence of age on the response to renal parenchymal loss, Yale J BioI Med 51: 341 (1978). 25. Claesson G: Experimental hydronephrosis in newborn rats: effects on renal morphology and function, Kongl Carolinska Medico Chir Inst Stockholm, 1987. 26. Laufer 1. and Griscom :--JT: Compensatory renal hvpertrophy: absence in utero and development in early life, AJR 113: 464 (1971).
437
27. Chevalier RL, Campbell F, and Brenbridge ANAG: Nephrosonography and renal scintigraphy in evaluation of newborn with renomegaly, Urology 24: 96 (1984). 28. Aperia A, et aI: Renal growth and function in patients nephrectomized in childhood, Acta Pediatr Scand 66: 185 (1977). 29. Claesson I, et aI: Compensatory kidney growth in children with urinary tract infection and unilateral renal scarring: an epidemiologic study, Kidney Int 20: 759 (1981). 30. Claesson I, and Jacobsson B: Standardisierte messungen der nierenparenchymdicke. Methode und diagnosticher wert bei kindem, in Olbing H (Ed): Rezidivierende nicht obstruktive harnwegserkrankungen bei kindem, Heidelberg, SpringerVerlag, 1980, pp 65-75. 31. Wilton P, et aI: Renal compensatory hypertrophy in children with unilateral renal disease, Acta Paediatr 69: 83 (1980). 32. King LR, The management of multicystic kidney and ureteropelvic junction obstruction, in King LR (Ed): Urologic Sur-
gery in Neonates and Young Infants, Philadelphia, WB Saunders, chap 9, 1988, pp 140-154. 33. Flake AW, et aI: Ureteropelvic junction obstruction in the fetus, J Pediatr Surg 21: 1058 (1986). 34. Thorup J, et al: The prognosis of surgically treated congenital hydronephrosis after diagnosis in utero, J Urol 134: 914 (1985). 35. Hanna MK, and Gluck R: Ureteropelvic junction obstruction during the first year of life, Urology 31: 41 (1988). 36. Wolpert JJ, Woodward JR, and Parrott TS: Pyeloplasty in the young infant, presented at Section on Pediatric Urology, American Academy of Pediatrics, San Francisco, October 16, 1988. 37. Koyle MA, and Ehrlich RM: Management of ureteropelvic junction obstruction in neonate, Urology 31: 496 (1988). 38. King LR, et aI: The case for immediate pyeloplasty in the neonate with upJ obstruction, J Uro1132: 725 (1984).
438
UROLOGY
I
MAY 1990
VOLUME XXXV, NUMBEH 5
