- Email: [email protected]
Ureteropelvic Junction Obstruction Associated with Ureteral Hypoplasia
0022-534 7/89/1422-0353$02.00/G THE J-OURNAL OP UROLOGY Copyright© 1989 by AMERICAN UROLOGICAL ASSOCIATION, INC,
VoL 142, August
Printed in U, S, A.
URETEROPELVIC JUNCTION OBSTRUCTION ASSOCIATED WITH URETERAL HYPOPLASIA TERRY D. ALLEN
AND
DOUGLAS A. HUSMANN
From the Division of Urology, University of Texas Southwestern Medical Center and Children's Medical Center, Dallas, Texas
ABSTRACT
We present 3 cases in which a neonate had unilateral hydronephrosis associated with a ureter hypoplastic throughout its length. In all 3 infants the affected kidney failed to visualize on excretory urography and function of the kidney proved to be disappointing despite its relatively normal appearance on external examination. In each case significant renal dysplasia was present microscopically and ultimately the kidney had to be removed. A retrograde pyelogram should be considered whenever a hydronephrotic kidney in an infant fails to visualize, and if this shows a long hypoplastic ureter that cannot be bypassed easily associated renal dysplasia should be suspected and consideration given to nephrectomy rather than pyeloplasty as the primary treatment modality. (J. Ural., 142: 353-355, 1989) Hydronephrosis secondary to ureteropelvic junction obstruction is a con1mon finding in pediatric urology, even more so now that prenatal sonography is being used so routinely in the care of the obstetrical patient. In the treatment of this disorder in the neonate the general philosophy has been to relieve the obstruction without great regard for the superficial appearance of the kidney, since experience has shown that a renal-sparing operation yields good results in most cases. 1 In the course of treating such patients, however, we have encountered a subset of children with hydronephrosis secondary to ureteropelvic junction obstruction in whom the outcome has been uniformly bad. Recognition of these patients in advance may influence the attending physician to alter the traditional conservative approach to the management of hydronephrosis in these specific instances. CASE REPORT
Case 1. B. D., a male Latin American neonate, was first discovered to have a hydronephrotic right kidney on prenatal ultrasonography at 32 weeks of gestation. After spontaneous delivery at 38 weeks the patient was noted to have a palpable right flank mass. Abdominal ultrasonography revealed findings consistent with a hydronephrotic kidney secondary to obstruction at the ureteropelvic junction. The left kidney appeared to be normal ultrasound and excretory urography (IVP) but there was no visible excretion of contrast material from the right even after 3 hours of observation. A 99 mtechnetiurn-diethylenetriaminepentaacetic acid (99 mTcrenal scan showed a rim of functioning tissue on the right side surrounding a central photopenic zone, and suggested that less than 10 per cent of the total renal function came from the right side. Renal exploration when the patient was 1 week old revealed a hydronephrotic kidney with an average parenchymal thickness of about 5 mm. The ureter below the ureteropelvic junction was atretic throughout its exposed length. With lacrimal duct probes the ureter was dilated to 3F but attempts to place a 2.8F silicone ureteral stent failed and, finally, a dismembered Anderson-Hynes pyeloplasty was done without a stent or nephrostomy tube. Despite a benign clinical course a 6-week followup examination by renal ultrasonography and 99 mTc-DTPA renal scan indicated that the kidney still was hydronephrotic and obstructed. Percutaneous puncture of the kidney confirmed the obstruction and revealed that urine output from this kidney Accepted for publication January 20, 1989.
353
amounted to only about 5 per cent of the total. Right nephrectomy was performed. Histological examination of the kidney revealed 90 per cent of the glomeruli to be involved with segmental or global sclerosis (fig. 1). In addition, there was periglomerular fibrosis and peritubular fibrosis suggestive of dysplasia although typical dysplastic features were absent. Case 2. M. T., a white female neonate, was first noted to have findings suggesting right hydronephrosis during routine prenatal sonography of the mother. After delivery renal sonography confirmed the presence of hydronephrosis consistent with congenital ureteropelvic junction obstruction. However, sufficient renal parenchyma was noted suggesting that the kidney was salvageable. A voiding cystourethrogram showed no vesicoureteral reflux and the serum creatinine was 0.3 mg./dl. (normal 0.3 to 0.6). An IVP showed a normal left kidney but no visible excretion of contrast material from the right side despite delayed films taken during a 4-hour period (fig. 2, A). A 99 mTc- DTP A nuclear scan revealed a normal left kidney and a thin rim of radioactivity surrounding an empty central zone on the right side (fig. 2, B). Preoperatively, a retrograde pyelogram revealed a string-like ureter that accepted a 3F catheter only in its lower third (fig. 2, C). At operation the renal parenchyma was thin but healthy in appearance, and draped around a dilated and tense renal pelvis obstructed at the ureteropelvic junction. The ureter was hypoand so narrow that even after dilation with a lacrimal probe it would accept only a 2.8F silicone catheter for a short distance. An Anderson-Hynes dismembered pyeloplasty was done over the silicone stent using interrupted 6-zero chromic sutures. A 12F nephrostomy tube was left indwelling for drainage but only about 15 per cent of the over-all urine output of the child ever came from the right kidney. During the postoperative period the silicone stent was injected with saline in an effort to dilate the ureter and promote peristaltic activity but this proved unsuccessful. The stent was removed in the office 13 days postoperatively at which time antegrade nephrostography showed no drainage through the repair site. The following day the child spiked a fever and was hospitalized with a Pseudomonas aeruginosa urinary tract infection. Further efforts to salvage the kidney were abandoned and right nephrectomy was done. Pathological examination of the kidney revealed hydronephrosis with acute and chronic pyelonephritis, and medullary dysplasia. Case 3. K. B., a white female neonate, was first discovered to have a hydronephrotic left kidney during prenatal ultrasonography at 30 weeks of gestation. Four weeks later, because of decreased fetal movements, an emergency cesarean section was
354
ALLEN AND HUSMANN
performed. After treatment for hypertension and pulmonary insufficiency, urological evaluation was requested. A voiding cystourethrogram revealed grade III/V vesicoureteral reflux on the right side but none on the left side. Abdominal ultrasound findings were consistent with left hydronephrosis secondary to obstruction at the ureteropelvic junction but suggested that a reasonable thickness of cortex was present. Subsequently, an IVP showed a normal right kidney but no excretion of contrast material from the left kidney, and a 99mTc-DTPA renal scan was interpreted as showing left hydronephrosis secondary to obstruction at the ureteropelvic junction with less than 10 per cent of over-all renal function provided by the left kidney. Two attempts to place a nephrostomy tube pe!"cutaneously in the hope of relieving the hypertension failed for technical reasons, and finally the left kidney was explored through the flank. The kidney was distended and tense but exhibited a renal parenchyma averaging 4 mm. thick. The ureter was hypoplastic throughout its exposed course in the flank and after excision of 3 cm. of ureter the remainder was dilated with lacrimal probes to 4F. An Anderson-Hynes dismembered pyeloplasty was done leaving a 3.5F silicone stent down the ureter and a lOF nephrostomy tube for drainage to the exterior. Biopsy of the kidney revealed segmental and global glomerulosclerosis with periglomerular and peritubular fibrosis, and marked tubular atrophy (fig. 3).
FIG. 1. Case 1. Microscopic examination of right kidney shows glomerulosclerosis. Reduced from XlOO.
Postoperatively, the patient remained hypertensive and urine output through the nephrostomy tube accounted for less than 3 per cent of the total urine output. After several weeks without improvement in the hypertension or renal function, and with inadequate drainage through to the bladder the kidney was removed. Within 3 weeks of nephrectomy the hypertension subsided and all antihypertensive medications were discontinued. DISCUSSION
There are certain elements common to all 3 cases. In the first place none of the 3 affected kidneys showed visible excretion of contrast material on an IVP even with delayed films. While the significance of this may be debated, given the immaturity of the kidney at this age and the presence of ureteral obstruction we find this to be somewhat unusual. In the absence of dysplasia the intrinsic functional capacity of the congenitally obstructed kidney generally is good in the neonate despite the degree of distension to which it may be subjected, and we have come to expect that such a kidney will provide about a third of the over-all renal function of the infant after full stabilization has occurred. Admittedly, the apparent function of the affected kidney in these patients as judged preoperatively by renal
FIG. 3. Case 3. Microscopic examination of left kidney shows glomerulosclerosis with periglomerular and peritubular fibrosis, and tubular atrophy. Reduced from XlOO.
FIG. 2. Case 2. A, IVP shows nonvisualization of right kidney. B, 99mTc scan demonstrates normal left kidney and thin rim of activity surrounding central zone on right side. C, right retrograde ureterogram through 3F ureteral catheter reveals hypoplastic ureter throughout its length. Arrow shows catheter tip.
URETEROPELVIC JUNCTWN OBSTRUCTION ASSOCIATED WITH URETERAL HYPOPLASIA
scintiscans was not but vve have found such studies to be somewhat unreliable in the neonate and are hesitant to make firm decisions on this basis alone. Obstruction in a patient of this age is more akin to an acute than a chronic obstruction and in an acutely obstructed kidney renal blood flow, impeded by the high intraluminal pressure present during the obstruction, may show dramatic recovery when the obstruction is relieved. We have seen infants with ureteropelvic junction obstruction whose scintiscans were reported as showing insignificant function of the affected kidney during the obstruction, only to find that after the obstruction was relieved the affected kidney accounted for more than 30 per cent of the over-all renal function. However, in our 3 cases the kidneys continued to perform poorly even after the obstruction was relieved. This was particularly disappointing in view of what appeared sonographically and on direct observation to be reasonable renal parenchyma. The explanation for this paradox seems to lie in the histological features of these kidneys, all 3 of which showed major intrinsic abnormalities, including glomerulosclerosis and renal dysplasia. It appears to be this rather than the obstruction per se that is the dominant factor in the poor function of these kidneys. In all 3 cases the ureter was thread-like and hypoplastic. Such urinary systems probably occupy a central position along the spectrum that runs between the more typical ureteropelvic junction obstruction on the one hand and multicystic renal dysplasia on the other. Hypoplasia of the ureter of this magnitude suggests that the ureter never had a major role in conducting urine to the bladder but rather represented an obstruction to urinary flow beginning at an early point in development. Indeed, the severity of the obstruction and/or its timing appears to be a significant factor in determining the quality of the associated kidney. Kidneys obstructed during the first half of fetal development, for example, tend to be dysplastic, while those obstructed later develop simple hydronephrosis.2· 3 In any event, the presence of extensive ureteral hypoplasia should alert the attending physician to the possibility of associated renal dysplasia of a magnitude sufficient to render pyeloplasty inappropriate. Frozen section biopsy of the kidney may be helpful before making a final decision in such cases but it appears that in most instances of this type nephrectomy will be the procedure of choice. In truth, none of these 3 kidneys was salvaged despite sometimes heroic efforts to do so. Dilating the ureter with a lacrimal probe, using tiny silicone stents and even injecting the stents with saline intermittently in the postoperative period all failed to establish a serviceable conduit to the bladder; the ultimate
355
outcome in each case was nephrectomy. One may speculate whether even more heroic efforts, such as the use of intestinal conduits, might have yielded a different outcome but when one considers the poor functional reserve of these kidneys to begin with, it would seem difficult to justify such an undertaking. Based on the experience gained in these 3 cases we now believe that the constellation of nonvisualization of the kidney on an IVP and poor function by renal scintiscan in the face of a poorly developed ureter portends significant renal dysplasia despite the external appearance of the kidney. Thus, whenever a unilateral hydronephrotic kidney fails to visualize on an IVP a retrograde pyelogram should be done, and if this reveals a tiny hypoplastic ureter that cannot be bypassed with a rotated or spiral flap a frozen section biopsy of the kidney should be performed. If biopsy confirms renal dysplasia we would recommend that the kidney be removed at the outset to spare the child what appears otherwise to be an inevitable second operation. REFERENCES 1. Kelalis, P. P.: Ureteropelvicjunction. In: Clinical Pediatric Urology,
2nd ed. Edited by P. P. Kelalis, L. R. King and A. B. Belman. Philadelphia: W. B. Saunders Co., chapt. 16, p. 464, 1985. 2. Beck, A. D.: The effect of intra-uterine urinary obstruction upon the development of the fetal kidney. J. Ural., 105: 784, 1971. 3. Glick, P. L., Harrison, M. R., Noall, R. A. and Villa, R. L.: Correction of congenital hydronephrosis in utero HI. Early mid-trimester ureteral obstruction produces renal dysplasia. J. Ped. Surg., 18: 681, 1983.
EDITORIAL COMMENT These 3 cases harbor a common but important theme. The constellation of some degree of parenchymal thickness on ultrasound, poor function on renal scan, nonvisualization on an IVP (with delayed films up to 24 hours) and atresia of most of the ureter al length is predictive of renal dysplasia. When renal parenchyma is dysplastic there is no reasonable hope that resolution of associated obstruction will lead to return of function. Therefore, nephrectomy is the best surgical option for these patients. Recognition of a long segment of meteral atresia is the clue that dysplasia is likely to be found histologically. Although routine retrograde pyelography has fallen into disfavor we use it in patients with presumed ureteropelvic junction obstruction to demonstrate the ureter distal to an obstruction (just before repair) when the ureter has not been visualized by some other modality.
George W. Kaplan Pediatric Urological Associates San Diego, California
VoL 142, August
Printed in U, S, A.
URETEROPELVIC JUNCTION OBSTRUCTION ASSOCIATED WITH URETERAL HYPOPLASIA TERRY D. ALLEN
AND
DOUGLAS A. HUSMANN
From the Division of Urology, University of Texas Southwestern Medical Center and Children's Medical Center, Dallas, Texas
ABSTRACT
We present 3 cases in which a neonate had unilateral hydronephrosis associated with a ureter hypoplastic throughout its length. In all 3 infants the affected kidney failed to visualize on excretory urography and function of the kidney proved to be disappointing despite its relatively normal appearance on external examination. In each case significant renal dysplasia was present microscopically and ultimately the kidney had to be removed. A retrograde pyelogram should be considered whenever a hydronephrotic kidney in an infant fails to visualize, and if this shows a long hypoplastic ureter that cannot be bypassed easily associated renal dysplasia should be suspected and consideration given to nephrectomy rather than pyeloplasty as the primary treatment modality. (J. Ural., 142: 353-355, 1989) Hydronephrosis secondary to ureteropelvic junction obstruction is a con1mon finding in pediatric urology, even more so now that prenatal sonography is being used so routinely in the care of the obstetrical patient. In the treatment of this disorder in the neonate the general philosophy has been to relieve the obstruction without great regard for the superficial appearance of the kidney, since experience has shown that a renal-sparing operation yields good results in most cases. 1 In the course of treating such patients, however, we have encountered a subset of children with hydronephrosis secondary to ureteropelvic junction obstruction in whom the outcome has been uniformly bad. Recognition of these patients in advance may influence the attending physician to alter the traditional conservative approach to the management of hydronephrosis in these specific instances. CASE REPORT
Case 1. B. D., a male Latin American neonate, was first discovered to have a hydronephrotic right kidney on prenatal ultrasonography at 32 weeks of gestation. After spontaneous delivery at 38 weeks the patient was noted to have a palpable right flank mass. Abdominal ultrasonography revealed findings consistent with a hydronephrotic kidney secondary to obstruction at the ureteropelvic junction. The left kidney appeared to be normal ultrasound and excretory urography (IVP) but there was no visible excretion of contrast material from the right even after 3 hours of observation. A 99 mtechnetiurn-diethylenetriaminepentaacetic acid (99 mTcrenal scan showed a rim of functioning tissue on the right side surrounding a central photopenic zone, and suggested that less than 10 per cent of the total renal function came from the right side. Renal exploration when the patient was 1 week old revealed a hydronephrotic kidney with an average parenchymal thickness of about 5 mm. The ureter below the ureteropelvic junction was atretic throughout its exposed length. With lacrimal duct probes the ureter was dilated to 3F but attempts to place a 2.8F silicone ureteral stent failed and, finally, a dismembered Anderson-Hynes pyeloplasty was done without a stent or nephrostomy tube. Despite a benign clinical course a 6-week followup examination by renal ultrasonography and 99 mTc-DTPA renal scan indicated that the kidney still was hydronephrotic and obstructed. Percutaneous puncture of the kidney confirmed the obstruction and revealed that urine output from this kidney Accepted for publication January 20, 1989.
353
amounted to only about 5 per cent of the total. Right nephrectomy was performed. Histological examination of the kidney revealed 90 per cent of the glomeruli to be involved with segmental or global sclerosis (fig. 1). In addition, there was periglomerular fibrosis and peritubular fibrosis suggestive of dysplasia although typical dysplastic features were absent. Case 2. M. T., a white female neonate, was first noted to have findings suggesting right hydronephrosis during routine prenatal sonography of the mother. After delivery renal sonography confirmed the presence of hydronephrosis consistent with congenital ureteropelvic junction obstruction. However, sufficient renal parenchyma was noted suggesting that the kidney was salvageable. A voiding cystourethrogram showed no vesicoureteral reflux and the serum creatinine was 0.3 mg./dl. (normal 0.3 to 0.6). An IVP showed a normal left kidney but no visible excretion of contrast material from the right side despite delayed films taken during a 4-hour period (fig. 2, A). A 99 mTc- DTP A nuclear scan revealed a normal left kidney and a thin rim of radioactivity surrounding an empty central zone on the right side (fig. 2, B). Preoperatively, a retrograde pyelogram revealed a string-like ureter that accepted a 3F catheter only in its lower third (fig. 2, C). At operation the renal parenchyma was thin but healthy in appearance, and draped around a dilated and tense renal pelvis obstructed at the ureteropelvic junction. The ureter was hypoand so narrow that even after dilation with a lacrimal probe it would accept only a 2.8F silicone catheter for a short distance. An Anderson-Hynes dismembered pyeloplasty was done over the silicone stent using interrupted 6-zero chromic sutures. A 12F nephrostomy tube was left indwelling for drainage but only about 15 per cent of the over-all urine output of the child ever came from the right kidney. During the postoperative period the silicone stent was injected with saline in an effort to dilate the ureter and promote peristaltic activity but this proved unsuccessful. The stent was removed in the office 13 days postoperatively at which time antegrade nephrostography showed no drainage through the repair site. The following day the child spiked a fever and was hospitalized with a Pseudomonas aeruginosa urinary tract infection. Further efforts to salvage the kidney were abandoned and right nephrectomy was done. Pathological examination of the kidney revealed hydronephrosis with acute and chronic pyelonephritis, and medullary dysplasia. Case 3. K. B., a white female neonate, was first discovered to have a hydronephrotic left kidney during prenatal ultrasonography at 30 weeks of gestation. Four weeks later, because of decreased fetal movements, an emergency cesarean section was
354
ALLEN AND HUSMANN
performed. After treatment for hypertension and pulmonary insufficiency, urological evaluation was requested. A voiding cystourethrogram revealed grade III/V vesicoureteral reflux on the right side but none on the left side. Abdominal ultrasound findings were consistent with left hydronephrosis secondary to obstruction at the ureteropelvic junction but suggested that a reasonable thickness of cortex was present. Subsequently, an IVP showed a normal right kidney but no excretion of contrast material from the left kidney, and a 99mTc-DTPA renal scan was interpreted as showing left hydronephrosis secondary to obstruction at the ureteropelvic junction with less than 10 per cent of over-all renal function provided by the left kidney. Two attempts to place a nephrostomy tube pe!"cutaneously in the hope of relieving the hypertension failed for technical reasons, and finally the left kidney was explored through the flank. The kidney was distended and tense but exhibited a renal parenchyma averaging 4 mm. thick. The ureter was hypoplastic throughout its exposed course in the flank and after excision of 3 cm. of ureter the remainder was dilated with lacrimal probes to 4F. An Anderson-Hynes dismembered pyeloplasty was done leaving a 3.5F silicone stent down the ureter and a lOF nephrostomy tube for drainage to the exterior. Biopsy of the kidney revealed segmental and global glomerulosclerosis with periglomerular and peritubular fibrosis, and marked tubular atrophy (fig. 3).
FIG. 1. Case 1. Microscopic examination of right kidney shows glomerulosclerosis. Reduced from XlOO.
Postoperatively, the patient remained hypertensive and urine output through the nephrostomy tube accounted for less than 3 per cent of the total urine output. After several weeks without improvement in the hypertension or renal function, and with inadequate drainage through to the bladder the kidney was removed. Within 3 weeks of nephrectomy the hypertension subsided and all antihypertensive medications were discontinued. DISCUSSION
There are certain elements common to all 3 cases. In the first place none of the 3 affected kidneys showed visible excretion of contrast material on an IVP even with delayed films. While the significance of this may be debated, given the immaturity of the kidney at this age and the presence of ureteral obstruction we find this to be somewhat unusual. In the absence of dysplasia the intrinsic functional capacity of the congenitally obstructed kidney generally is good in the neonate despite the degree of distension to which it may be subjected, and we have come to expect that such a kidney will provide about a third of the over-all renal function of the infant after full stabilization has occurred. Admittedly, the apparent function of the affected kidney in these patients as judged preoperatively by renal
FIG. 3. Case 3. Microscopic examination of left kidney shows glomerulosclerosis with periglomerular and peritubular fibrosis, and tubular atrophy. Reduced from XlOO.
FIG. 2. Case 2. A, IVP shows nonvisualization of right kidney. B, 99mTc scan demonstrates normal left kidney and thin rim of activity surrounding central zone on right side. C, right retrograde ureterogram through 3F ureteral catheter reveals hypoplastic ureter throughout its length. Arrow shows catheter tip.
URETEROPELVIC JUNCTWN OBSTRUCTION ASSOCIATED WITH URETERAL HYPOPLASIA
scintiscans was not but vve have found such studies to be somewhat unreliable in the neonate and are hesitant to make firm decisions on this basis alone. Obstruction in a patient of this age is more akin to an acute than a chronic obstruction and in an acutely obstructed kidney renal blood flow, impeded by the high intraluminal pressure present during the obstruction, may show dramatic recovery when the obstruction is relieved. We have seen infants with ureteropelvic junction obstruction whose scintiscans were reported as showing insignificant function of the affected kidney during the obstruction, only to find that after the obstruction was relieved the affected kidney accounted for more than 30 per cent of the over-all renal function. However, in our 3 cases the kidneys continued to perform poorly even after the obstruction was relieved. This was particularly disappointing in view of what appeared sonographically and on direct observation to be reasonable renal parenchyma. The explanation for this paradox seems to lie in the histological features of these kidneys, all 3 of which showed major intrinsic abnormalities, including glomerulosclerosis and renal dysplasia. It appears to be this rather than the obstruction per se that is the dominant factor in the poor function of these kidneys. In all 3 cases the ureter was thread-like and hypoplastic. Such urinary systems probably occupy a central position along the spectrum that runs between the more typical ureteropelvic junction obstruction on the one hand and multicystic renal dysplasia on the other. Hypoplasia of the ureter of this magnitude suggests that the ureter never had a major role in conducting urine to the bladder but rather represented an obstruction to urinary flow beginning at an early point in development. Indeed, the severity of the obstruction and/or its timing appears to be a significant factor in determining the quality of the associated kidney. Kidneys obstructed during the first half of fetal development, for example, tend to be dysplastic, while those obstructed later develop simple hydronephrosis.2· 3 In any event, the presence of extensive ureteral hypoplasia should alert the attending physician to the possibility of associated renal dysplasia of a magnitude sufficient to render pyeloplasty inappropriate. Frozen section biopsy of the kidney may be helpful before making a final decision in such cases but it appears that in most instances of this type nephrectomy will be the procedure of choice. In truth, none of these 3 kidneys was salvaged despite sometimes heroic efforts to do so. Dilating the ureter with a lacrimal probe, using tiny silicone stents and even injecting the stents with saline intermittently in the postoperative period all failed to establish a serviceable conduit to the bladder; the ultimate
355
outcome in each case was nephrectomy. One may speculate whether even more heroic efforts, such as the use of intestinal conduits, might have yielded a different outcome but when one considers the poor functional reserve of these kidneys to begin with, it would seem difficult to justify such an undertaking. Based on the experience gained in these 3 cases we now believe that the constellation of nonvisualization of the kidney on an IVP and poor function by renal scintiscan in the face of a poorly developed ureter portends significant renal dysplasia despite the external appearance of the kidney. Thus, whenever a unilateral hydronephrotic kidney fails to visualize on an IVP a retrograde pyelogram should be done, and if this reveals a tiny hypoplastic ureter that cannot be bypassed with a rotated or spiral flap a frozen section biopsy of the kidney should be performed. If biopsy confirms renal dysplasia we would recommend that the kidney be removed at the outset to spare the child what appears otherwise to be an inevitable second operation. REFERENCES 1. Kelalis, P. P.: Ureteropelvicjunction. In: Clinical Pediatric Urology,
2nd ed. Edited by P. P. Kelalis, L. R. King and A. B. Belman. Philadelphia: W. B. Saunders Co., chapt. 16, p. 464, 1985. 2. Beck, A. D.: The effect of intra-uterine urinary obstruction upon the development of the fetal kidney. J. Ural., 105: 784, 1971. 3. Glick, P. L., Harrison, M. R., Noall, R. A. and Villa, R. L.: Correction of congenital hydronephrosis in utero HI. Early mid-trimester ureteral obstruction produces renal dysplasia. J. Ped. Surg., 18: 681, 1983.
EDITORIAL COMMENT These 3 cases harbor a common but important theme. The constellation of some degree of parenchymal thickness on ultrasound, poor function on renal scan, nonvisualization on an IVP (with delayed films up to 24 hours) and atresia of most of the ureter al length is predictive of renal dysplasia. When renal parenchyma is dysplastic there is no reasonable hope that resolution of associated obstruction will lead to return of function. Therefore, nephrectomy is the best surgical option for these patients. Recognition of a long segment of meteral atresia is the clue that dysplasia is likely to be found histologically. Although routine retrograde pyelography has fallen into disfavor we use it in patients with presumed ureteropelvic junction obstruction to demonstrate the ureter distal to an obstruction (just before repair) when the ureter has not been visualized by some other modality.
George W. Kaplan Pediatric Urological Associates San Diego, California