- Email: [email protected]
Role of Urinary Transforming Growth Factor-β1 Concentration in the Diagnosis of Upper Urinary Tract Obstruction in Children
0022-5347/04/1684-1798/0 THE JOURNAL OF UROLOGY® Copyright © 2004 by AMERICAN UROLOGICAL ASSOCIATION, INC.®
Vol. 168, 1798 –1800, October 2002 Printed in U.S.A.
DOI: 10.1097/01.ju.0000027231.84450.8f
ROLE OF URINARY TRANSFORMING GROWTH FACTOR-1 CONCENTRATION IN THE DIAGNOSIS OF UPPER URINARY TRACT OBSTRUCTION IN CHILDREN M. T. EL-SHERBINY, O. M. MOUSA, A. A. SHOKEIR
AND
M. A. GHONEIM
From The Urology and Nephrology Center, Mansoura University, Mansoura, Egypt
ABSTRACT
Purpose: We evaluate the role of transforming growth factor-1 (TGF-1) in the diagnosis and followup of children with pelviureteral junction obstruction. Materials and Methods: TGF-1 concentration was measured in renal pelvic and bladder urine samples obtained from 15 children who underwent surgery for symptomatic unilateral pelviureteral junction obstruction. Bladder urine TGF-1 was also measured in 11 age matched patients with dilated nonobstructed kidneys as documented by nonobstructed washout curve and half-time drainage less than 10 minutes on diuretic renography. In the obstructed group bladder urine TGF-1 was measured 3 months after surgery. Results: In the obstructed group mean TGF-1 plus or minus SD in the renal pelvic urine was 285 ⫾ 191 pg./mg. creatinine, or 4-fold that of bladder urine (p ⬎0.001). Mean bladder urine TGF-1 was 3-fold higher in patients with upper tract obstruction than in controls (68 ⫾ 59 versus 22 ⫾ 18 pg./mg. creatinine, p ⬍0.003). Mean bladder TGF-1 3 months after surgery showed a trend towards a decrease, albeit still insignificant (68 ⫾ 59 versus 39 ⫾ 31 pg./mg. creatinine for preoperatively versus postoperatively, p ⬍0.08). Using a bladder urine concentration of 29 pg./mg. creatinine as a cutoff between obstruction and no obstruction, TGF-1 was 80% sensitive, 82% specific and 81% accurate for the diagnosis of obstruction. Conclusions: Bladder urine TGF-1 is a useful noninvasive tool for diagnosis of upper urinary tract obstruction. At 3 months following corrective surgery there is a trend towards decrease in bladder TGF-1 concentration in comparison to the preoperative value. KEY WORDS: bladder, transforming growth factor beta; hydronephrosis, kidney, ureteral obstruction
Pelviureteral junction obstruction is the most common cause of hydronephrosis in children, and the diagnosis of obstruction versus nonobstruction is difficult. The Whitaker test is considered by some to be the gold standard for diagnosis of obstruction but it is invasive and has the potential of false-positive results. Therefore, it has not gained wide use particularly in children. Today diuretic renography is the most widely used method to evaluate renal function and drainage.1 However, the test remains invasive, using ionizing irradiation with surrounding controversy especially in infants and younger children.2 The noninvasive nature of measurement of renal resistive index is appealing in its potential application for diagnosis of obstruction. However, resistive index is age dependent and its potential usefulness in children is yet to be determined.3 The availability of a noninvasive test to predict or aid in identifying which cases require surgery for pelviureteral junction obstruction before the loss of differential renal function would be valuable. Palmer et al have shown that transforming growth factor-1 (TGF-1) is detectable in the urine in cases of normal and pathological conditions of the urinary tract, specifically vesicoureteral reflux and hydronephrosis.4 Furness et al demonstrated that bladder urine TGF-1 was significantly increased in children with upper urinary tract obstruction compared to that of controls.5 They were the first to identify a bladder urinary marker that correlates with upper urinary tract obstruction with greater than 90% sensitivity. However, their study was criticized by the lack of correlation to patients with dilated nonobstructed uropathy treated nonoperatively. Furthermore, the study did not correlate between preoperative and postoperative TGF-1 values. We correlated urinary TGF-1 concentrations in 2 groups
of patients with dilated obstructed and dilated nonobstructed uropathy. The obstructed group was treated surgically while the nonobstructed group was followed. Values of TGF-1 before and after surgical treatment of the obstructed group were compared with those of the nonobstructed group. PATIENTS AND METHODS
The study group comprised 8 boys and 7 girls (15 renal units) with a mean age plus or minus SD of 5.2 ⫾ 4.7 years (range 4 months to 14 years). All patients were diagnosed with unilateral pelviureteral junction obstruction and pyeloplasty was performed. In all children the contralateral kidneys were essentially normal. The initial presentation was a flank mass in 4 cases, pain in 7 and history of urinary tract infection in 4. Hydronephrosis was graded according to the Society for Fetal Urology classification,6 and was grade 3 or greater in all cases. Upper urinary tract obstruction was diagnosed in 15 children when well-tempered renography revealed a prolonged half-time drainage (20 minutes or greater) and/or an obstructive washout curve pattern during the diuretic phase. All children had sterile urine before surgery and none had concurrent vesicoureteral reflux. After induction of anesthesia, urine samples were obtained from the bladder by a urethral catheter and from the renal pelvis by needle aspiration. At 3 months postoperatively bladder urine TGF-1 was measured and a renogram was repeated. The study group was compared to a control group of 11 age matched children (10 boys and 1 girl, mean age 5.4 ⫾ 4.2, range 0.6 to 12) with unilateral dilated nonobstructed kidneys as documented by nonobstructed washout curve and half-time drainage less than 10 minutes on diuretic renography. Four children were diagnosed with prenatal ultrasound
1798
TRANSFORMING GROWTH FACTOR-1 AND URINARY TRACT OBSTRUCTION
and treated conservatively. The remaining 7 children had residual unilateral nonsymptomatic, nonobstructive dilatation that persisted for more than 1 year following successful pyeloplasty. All children in the control group had grade 3 or greater hydronephrosis and none had concurrent vesicoureteral reflux or urinary tract infection. Bladder urine was obtained from all children in this group. TGF-1 was measured using the enzyme-linked immunoassay method similar to that described by Furness et al.5 Statistical analysis was performed using the Mann-Whitney U and Wilcoxon rank sum tests, with p ⬍0.05 considered significant. RESULTS
In the obstructed group mean TGF-1 plus or minus SD in the renal pelvic urine was 285 ⫾ 191 pg./mg. creatinine, or 4-fold that of the bladder urine (p ⬎0.001). Mean bladder urine TGF-1 was 3-fold higher in patients with upper tract obstruction than in controls (68 ⫾ 59 versus 22 ⫾ 18 pg./mg. creatinine, respectively, p ⬍0.003). In the obstructed group changes in split renal function did not exceed 5% at 3 months postoperatively. Mean bladder TGF-1 3 months after surgery showed a trend towards a decrease, albeit still insignificant (68 ⫾ 59 versus 39 ⫾ 31 pg./mg. creatinine preoperatively versus postoperatively, p ⬍0.08). The figure shows the mean urinary TGF-1 concentrations of patients and controls. ROC curve was drawn to identify the cutoff value of the urinary TGF-1 concentration that provided the highest sensitivity and specificity. Using a bladder urine concentration of 29 pg./mg. creatinine as a cutoff between obstruction and no obstruction, TGF-1 concentration was 80% sensitive, 82% specific and 81% accurate for the diagnosis of obstruction (see table). We assessed the possible influence of chronological age on urinary TGF-1 by comparing urinary TGF-1 concentration in children younger than 1 year versus older children.7 In the study and control groups mean bladder urine TGF-1 concentration in children younger than 1 year was not significantly different from values in older children. In the obstructed group mean renal pelvic urine TGF-1 concentration was higher in children younger than 1 year than in older children (465 versus 218 pg./mg. creatinine, p ⬍0.04). A possible influence of split renal function on the level of urinary TGF-1 was also evaluated. There was no significant linear correlation of TGF-1 with the split renographic clearance in either the obstructed or control groups.
Mean TGF-1 concentrations (Conc) in renal pelvis and bladder of children with pelviureteral junction obstruction, and in bladder 3 months after pyeloplasty versus controls, as corrected for creatinine. Bladder urine TGF-1 in obstructed group was 3-fold that in control group. In obstructed group mean renal pelvic urine TGF-1 was 4-fold that in bladder urine. Bladder urine TGF-1 3-months after pyeloplasty showed trend toward decrease.
1799
Agreement between TGF-1 and diuretic renography results in 26 patients and controls Renogram Results
Predicted TGF Results*
Pos.
Pos. 12 Neg. 3 * Sensitivity 80%, specificity 82%, overall accuracy 81%.
Neg. 2 9
DISCUSSION
Increasing evidence suggests that in a variety of pathological states including ureteral obstruction, excessive renal TGF-1 production is responsible for dysregulation of extracellular matrix production and the development of progressive renal fibrosis.7–10 Moreover, Isaka et al recently demonstrated that interstitial fibrosis in unilateral ureteral obstruction could be blocked with TGF-1 antisense oligodeoxynucleotides.11 Palmer et al demonstrated that increased renal tissue levels of TGF-1 in response to urinary tract obstruction were associated with an increase of urinary TGF-1.4 Furness et al reported that bladder urine TGF-1 was significantly increased in children with upper urinary tract obstruction compared to normal controls.5 We demonstrated that bladder urine TGF-1 in children with upper urinary tract obstruction was significantly increased 3-fold that of children with dilated nonobstructed kidneys. In fact, the sensitivity of the test in our study was lower than that reported by Furness et al (80% versus 92%, respectively), which could be attributed to the difference in the control group in both studies. Furness et al included normal children without upper tract dilatation in the control group, which is in contrast to our control group of children with a dilated nonobstructed upper tract. Thus, the 80% sensitivity determined in our study is more practical, since the question of obstruction versus no obstruction will be raised only in the presence of dilatation. These data suggest that increased bladder urine TGF-1 may be combined with other useful diagnostic tests to differentiate between dilated obstructed and dilated nonobstructed kidneys, specifically in children with grade 3 or greater hydronephrosis. These combinations may allow the development of a scoring system for the diagnosis of obstruction. Mean urinary TGF-1 concentration in the renal pelvis and bladder of our obstructed group is lower than that reported by Furness et al5 renal pelvis 285 versus 378, and bladder 68 versus 195 pg./mg. creatinine. The cutoff value of bladder urine TGF-1 for the diagnosis of obstruction in our series was 29 pg./mg. creatinine, which was lower than the 61 pg./mg. creatinine reported by Furness et al. This difference could be explained by the difference in mean patient age, which was 5 years versus 5 months.5 Seremetis and Maizels studied TGF-1 expression in 15 specimens of obstructed pelves from children 6 weeks to 13 years old, and demonstrated that TGF-1 expression was significantly lower in children with chronic (after age 1 year) than those with acute (before age 1 year) presentation.7 Furness et al also showed that the mean bladder and renal pelvic urine TGF-1 was lower in older children than those younger than 2 years.5 Similarly we demonstrated that in the obstructed group the mean renal pelvic urine TGF-1 concentration was lower in older children than in those younger than 1 year (218 versus 465 pg./mg. creatinine, p ⬍0.04). These lower urinary TGF-1 concentrations in older children with obstruction may reflect a lower and more steady state of production of TGF-1 with long-standing obstruction. However, further research is needed to define the age dependency of urinary TGF-1 concentration. The level of TGF- expression has been previously correlated with the outcome after relief of obstruction. Seremetis and Maizels demonstrated that cases with higher mean level of TGF- expression had significant improvement in postop-
1800
TRANSFORMING GROWTH FACTOR-1 AND URINARY TRACT OBSTRUCTION
erative tests than those with lower expression.7 This observation was explained by the fact that the highest level of TGF- expression was derived from pelves of patients who presented before age 1 year when recoverability of renal function after relief of obstruction is highest. In our study the absent correlation between TGF-1 urinary concentrations and renal function could be explained partly by the relatively higher mean age and the small number of our patients. A study of the correlation between TGF-1 and renal function in a larger group of patients may help predict recoverability of renal function after relief of obstruction.12 To our knowledge, the correlation between urinary TGF-1 levels before and after surgical correction of obstruction has never been previously evaluated in a clinical setting. We have demonstrated that the mean bladder TGF-1 3 months after surgery shows a trend towards a decrease, albeit still insignificant (68 ⫾ 59 versus 39 ⫾ 31 pg./mg. creatinine preoperatively versus postoperatively, p ⬍0.08). Most likely, a significant difference will be observed with longer followup. These data suggest that determination of bladder urine concentration of TGF-1 may be also useful for followup after surgical correction of obstruction. However, longitudinal studies of a large patient population are needed to determine the reference range to diagnose obstruction and correlate urinary levels with final outcome. CONCLUSIONS
Bladder urine TGF-1 in patients with upper urinary tract obstruction is significantly increased compared with that in patients with dilated nonobstructed upper urinary tract (80% sensitive, 82% specific, 81% accurate). The increased bladder urine TGF-1 may be combined with other useful diagnostic tests to confirm clinically significant upper urinary tract obstruction. At followup there is a trend towards decrease of bladder TGF-1 concentration compared to the preoperative value, which may be useful for evaluating success after corrective surgery.
REFERENCES
1. Salem, Y. H., Majd, M., Rushton, H. G. et al: Outcome analysis of pediatric pyeloplasty as a function of patient age, presentation and differential renal function. J Urol, 154: 1889, 1995 2. Kass, E. J., Majd, M. and Belman, A. B.: Comparison of pediatric renogram and the pressure perfusion study in children. J Urol, 134: 92, 1985 3. Shokeir, A. A., Provoost, A. P., El-Azab, M. et al: Renal Doppler ultrasound in children with normal upper urinary tracts: effect of fasting, hydration with normal saline, and furosemide administration. Urology, 47: 740, 1996 4. Palmer, L. S., Maizels, M., Kaplan, W. E., Firlit, C. F. and Cheng, E. Y.: Urine levels of transforming growth factor-beta 1 in children with ureteropelvic junction obstruction. Urology, 50: 5, 1997 5. Furness, P. D., Maizels, M., Han, S. W. et al: Elevated bladder urine concentration of transforming growth factor-1 correlates with upper urinary tract obstruction in children. J Urol, 162:, 1033, 1999 6. Fernbach, S. K., Maizels, M. and Conway, J. J.: Ultrasound grading of hydronephrosis: introduction of the system used by the Society for Fetal Urology. Pediatr Radiol, 23: 478, 1993 7. Seremetis, G. M. and Maizels, M.: TGF-beta mRNA expression in the renal pelvis after experimental and clinical ureteropelvic junction obstruction. J Urol, 156: 261, 1996 8. Yamamoto, T., Noble, N. A., Miller, D. E. et al: Sustained expression of TGF-beta 1 underlies development of progressive kidney fibrosis. Kidney Int, 45: 916, 1994 9. Chung, K. H. and Chevalier, R. L.: Arrested development of the neonatal kidney following chronic ureteral obstruction. J Urol, 155: 1139, 1996 10. Seseke, F., Thelen, P., Heuser, M. et al: Impaired nephrogenesis in rats with congenital obstructive uropathy. J Urol, 165: 2289, 2001 11. Isaka, Y., Tsujie, M., Ando, Y., Nakamura, H. et al: Transforming growth factor-beta 1 antisense oligodeoxynucleotides block interstitial fibrosis in unilateral ureteral obstruction. Kidney Int, 58: 1885, 2000 12. Shokeir, A. A., Provoost, A. P. and Nijman, R. J. M.: Recoverability of renal function after relief of chronic partial upper urinary tract obstruction. BJU Int, 83: 11, 1999
Vol. 168, 1798 –1800, October 2002 Printed in U.S.A.
DOI: 10.1097/01.ju.0000027231.84450.8f
ROLE OF URINARY TRANSFORMING GROWTH FACTOR-1 CONCENTRATION IN THE DIAGNOSIS OF UPPER URINARY TRACT OBSTRUCTION IN CHILDREN M. T. EL-SHERBINY, O. M. MOUSA, A. A. SHOKEIR
AND
M. A. GHONEIM
From The Urology and Nephrology Center, Mansoura University, Mansoura, Egypt
ABSTRACT
Purpose: We evaluate the role of transforming growth factor-1 (TGF-1) in the diagnosis and followup of children with pelviureteral junction obstruction. Materials and Methods: TGF-1 concentration was measured in renal pelvic and bladder urine samples obtained from 15 children who underwent surgery for symptomatic unilateral pelviureteral junction obstruction. Bladder urine TGF-1 was also measured in 11 age matched patients with dilated nonobstructed kidneys as documented by nonobstructed washout curve and half-time drainage less than 10 minutes on diuretic renography. In the obstructed group bladder urine TGF-1 was measured 3 months after surgery. Results: In the obstructed group mean TGF-1 plus or minus SD in the renal pelvic urine was 285 ⫾ 191 pg./mg. creatinine, or 4-fold that of bladder urine (p ⬎0.001). Mean bladder urine TGF-1 was 3-fold higher in patients with upper tract obstruction than in controls (68 ⫾ 59 versus 22 ⫾ 18 pg./mg. creatinine, p ⬍0.003). Mean bladder TGF-1 3 months after surgery showed a trend towards a decrease, albeit still insignificant (68 ⫾ 59 versus 39 ⫾ 31 pg./mg. creatinine for preoperatively versus postoperatively, p ⬍0.08). Using a bladder urine concentration of 29 pg./mg. creatinine as a cutoff between obstruction and no obstruction, TGF-1 was 80% sensitive, 82% specific and 81% accurate for the diagnosis of obstruction. Conclusions: Bladder urine TGF-1 is a useful noninvasive tool for diagnosis of upper urinary tract obstruction. At 3 months following corrective surgery there is a trend towards decrease in bladder TGF-1 concentration in comparison to the preoperative value. KEY WORDS: bladder, transforming growth factor beta; hydronephrosis, kidney, ureteral obstruction
Pelviureteral junction obstruction is the most common cause of hydronephrosis in children, and the diagnosis of obstruction versus nonobstruction is difficult. The Whitaker test is considered by some to be the gold standard for diagnosis of obstruction but it is invasive and has the potential of false-positive results. Therefore, it has not gained wide use particularly in children. Today diuretic renography is the most widely used method to evaluate renal function and drainage.1 However, the test remains invasive, using ionizing irradiation with surrounding controversy especially in infants and younger children.2 The noninvasive nature of measurement of renal resistive index is appealing in its potential application for diagnosis of obstruction. However, resistive index is age dependent and its potential usefulness in children is yet to be determined.3 The availability of a noninvasive test to predict or aid in identifying which cases require surgery for pelviureteral junction obstruction before the loss of differential renal function would be valuable. Palmer et al have shown that transforming growth factor-1 (TGF-1) is detectable in the urine in cases of normal and pathological conditions of the urinary tract, specifically vesicoureteral reflux and hydronephrosis.4 Furness et al demonstrated that bladder urine TGF-1 was significantly increased in children with upper urinary tract obstruction compared to that of controls.5 They were the first to identify a bladder urinary marker that correlates with upper urinary tract obstruction with greater than 90% sensitivity. However, their study was criticized by the lack of correlation to patients with dilated nonobstructed uropathy treated nonoperatively. Furthermore, the study did not correlate between preoperative and postoperative TGF-1 values. We correlated urinary TGF-1 concentrations in 2 groups
of patients with dilated obstructed and dilated nonobstructed uropathy. The obstructed group was treated surgically while the nonobstructed group was followed. Values of TGF-1 before and after surgical treatment of the obstructed group were compared with those of the nonobstructed group. PATIENTS AND METHODS
The study group comprised 8 boys and 7 girls (15 renal units) with a mean age plus or minus SD of 5.2 ⫾ 4.7 years (range 4 months to 14 years). All patients were diagnosed with unilateral pelviureteral junction obstruction and pyeloplasty was performed. In all children the contralateral kidneys were essentially normal. The initial presentation was a flank mass in 4 cases, pain in 7 and history of urinary tract infection in 4. Hydronephrosis was graded according to the Society for Fetal Urology classification,6 and was grade 3 or greater in all cases. Upper urinary tract obstruction was diagnosed in 15 children when well-tempered renography revealed a prolonged half-time drainage (20 minutes or greater) and/or an obstructive washout curve pattern during the diuretic phase. All children had sterile urine before surgery and none had concurrent vesicoureteral reflux. After induction of anesthesia, urine samples were obtained from the bladder by a urethral catheter and from the renal pelvis by needle aspiration. At 3 months postoperatively bladder urine TGF-1 was measured and a renogram was repeated. The study group was compared to a control group of 11 age matched children (10 boys and 1 girl, mean age 5.4 ⫾ 4.2, range 0.6 to 12) with unilateral dilated nonobstructed kidneys as documented by nonobstructed washout curve and half-time drainage less than 10 minutes on diuretic renography. Four children were diagnosed with prenatal ultrasound
1798
TRANSFORMING GROWTH FACTOR-1 AND URINARY TRACT OBSTRUCTION
and treated conservatively. The remaining 7 children had residual unilateral nonsymptomatic, nonobstructive dilatation that persisted for more than 1 year following successful pyeloplasty. All children in the control group had grade 3 or greater hydronephrosis and none had concurrent vesicoureteral reflux or urinary tract infection. Bladder urine was obtained from all children in this group. TGF-1 was measured using the enzyme-linked immunoassay method similar to that described by Furness et al.5 Statistical analysis was performed using the Mann-Whitney U and Wilcoxon rank sum tests, with p ⬍0.05 considered significant. RESULTS
In the obstructed group mean TGF-1 plus or minus SD in the renal pelvic urine was 285 ⫾ 191 pg./mg. creatinine, or 4-fold that of the bladder urine (p ⬎0.001). Mean bladder urine TGF-1 was 3-fold higher in patients with upper tract obstruction than in controls (68 ⫾ 59 versus 22 ⫾ 18 pg./mg. creatinine, respectively, p ⬍0.003). In the obstructed group changes in split renal function did not exceed 5% at 3 months postoperatively. Mean bladder TGF-1 3 months after surgery showed a trend towards a decrease, albeit still insignificant (68 ⫾ 59 versus 39 ⫾ 31 pg./mg. creatinine preoperatively versus postoperatively, p ⬍0.08). The figure shows the mean urinary TGF-1 concentrations of patients and controls. ROC curve was drawn to identify the cutoff value of the urinary TGF-1 concentration that provided the highest sensitivity and specificity. Using a bladder urine concentration of 29 pg./mg. creatinine as a cutoff between obstruction and no obstruction, TGF-1 concentration was 80% sensitive, 82% specific and 81% accurate for the diagnosis of obstruction (see table). We assessed the possible influence of chronological age on urinary TGF-1 by comparing urinary TGF-1 concentration in children younger than 1 year versus older children.7 In the study and control groups mean bladder urine TGF-1 concentration in children younger than 1 year was not significantly different from values in older children. In the obstructed group mean renal pelvic urine TGF-1 concentration was higher in children younger than 1 year than in older children (465 versus 218 pg./mg. creatinine, p ⬍0.04). A possible influence of split renal function on the level of urinary TGF-1 was also evaluated. There was no significant linear correlation of TGF-1 with the split renographic clearance in either the obstructed or control groups.
Mean TGF-1 concentrations (Conc) in renal pelvis and bladder of children with pelviureteral junction obstruction, and in bladder 3 months after pyeloplasty versus controls, as corrected for creatinine. Bladder urine TGF-1 in obstructed group was 3-fold that in control group. In obstructed group mean renal pelvic urine TGF-1 was 4-fold that in bladder urine. Bladder urine TGF-1 3-months after pyeloplasty showed trend toward decrease.
1799
Agreement between TGF-1 and diuretic renography results in 26 patients and controls Renogram Results
Predicted TGF Results*
Pos.
Pos. 12 Neg. 3 * Sensitivity 80%, specificity 82%, overall accuracy 81%.
Neg. 2 9
DISCUSSION
Increasing evidence suggests that in a variety of pathological states including ureteral obstruction, excessive renal TGF-1 production is responsible for dysregulation of extracellular matrix production and the development of progressive renal fibrosis.7–10 Moreover, Isaka et al recently demonstrated that interstitial fibrosis in unilateral ureteral obstruction could be blocked with TGF-1 antisense oligodeoxynucleotides.11 Palmer et al demonstrated that increased renal tissue levels of TGF-1 in response to urinary tract obstruction were associated with an increase of urinary TGF-1.4 Furness et al reported that bladder urine TGF-1 was significantly increased in children with upper urinary tract obstruction compared to normal controls.5 We demonstrated that bladder urine TGF-1 in children with upper urinary tract obstruction was significantly increased 3-fold that of children with dilated nonobstructed kidneys. In fact, the sensitivity of the test in our study was lower than that reported by Furness et al (80% versus 92%, respectively), which could be attributed to the difference in the control group in both studies. Furness et al included normal children without upper tract dilatation in the control group, which is in contrast to our control group of children with a dilated nonobstructed upper tract. Thus, the 80% sensitivity determined in our study is more practical, since the question of obstruction versus no obstruction will be raised only in the presence of dilatation. These data suggest that increased bladder urine TGF-1 may be combined with other useful diagnostic tests to differentiate between dilated obstructed and dilated nonobstructed kidneys, specifically in children with grade 3 or greater hydronephrosis. These combinations may allow the development of a scoring system for the diagnosis of obstruction. Mean urinary TGF-1 concentration in the renal pelvis and bladder of our obstructed group is lower than that reported by Furness et al5 renal pelvis 285 versus 378, and bladder 68 versus 195 pg./mg. creatinine. The cutoff value of bladder urine TGF-1 for the diagnosis of obstruction in our series was 29 pg./mg. creatinine, which was lower than the 61 pg./mg. creatinine reported by Furness et al. This difference could be explained by the difference in mean patient age, which was 5 years versus 5 months.5 Seremetis and Maizels studied TGF-1 expression in 15 specimens of obstructed pelves from children 6 weeks to 13 years old, and demonstrated that TGF-1 expression was significantly lower in children with chronic (after age 1 year) than those with acute (before age 1 year) presentation.7 Furness et al also showed that the mean bladder and renal pelvic urine TGF-1 was lower in older children than those younger than 2 years.5 Similarly we demonstrated that in the obstructed group the mean renal pelvic urine TGF-1 concentration was lower in older children than in those younger than 1 year (218 versus 465 pg./mg. creatinine, p ⬍0.04). These lower urinary TGF-1 concentrations in older children with obstruction may reflect a lower and more steady state of production of TGF-1 with long-standing obstruction. However, further research is needed to define the age dependency of urinary TGF-1 concentration. The level of TGF- expression has been previously correlated with the outcome after relief of obstruction. Seremetis and Maizels demonstrated that cases with higher mean level of TGF- expression had significant improvement in postop-
1800
TRANSFORMING GROWTH FACTOR-1 AND URINARY TRACT OBSTRUCTION
erative tests than those with lower expression.7 This observation was explained by the fact that the highest level of TGF- expression was derived from pelves of patients who presented before age 1 year when recoverability of renal function after relief of obstruction is highest. In our study the absent correlation between TGF-1 urinary concentrations and renal function could be explained partly by the relatively higher mean age and the small number of our patients. A study of the correlation between TGF-1 and renal function in a larger group of patients may help predict recoverability of renal function after relief of obstruction.12 To our knowledge, the correlation between urinary TGF-1 levels before and after surgical correction of obstruction has never been previously evaluated in a clinical setting. We have demonstrated that the mean bladder TGF-1 3 months after surgery shows a trend towards a decrease, albeit still insignificant (68 ⫾ 59 versus 39 ⫾ 31 pg./mg. creatinine preoperatively versus postoperatively, p ⬍0.08). Most likely, a significant difference will be observed with longer followup. These data suggest that determination of bladder urine concentration of TGF-1 may be also useful for followup after surgical correction of obstruction. However, longitudinal studies of a large patient population are needed to determine the reference range to diagnose obstruction and correlate urinary levels with final outcome. CONCLUSIONS
Bladder urine TGF-1 in patients with upper urinary tract obstruction is significantly increased compared with that in patients with dilated nonobstructed upper urinary tract (80% sensitive, 82% specific, 81% accurate). The increased bladder urine TGF-1 may be combined with other useful diagnostic tests to confirm clinically significant upper urinary tract obstruction. At followup there is a trend towards decrease of bladder TGF-1 concentration compared to the preoperative value, which may be useful for evaluating success after corrective surgery.
REFERENCES
1. Salem, Y. H., Majd, M., Rushton, H. G. et al: Outcome analysis of pediatric pyeloplasty as a function of patient age, presentation and differential renal function. J Urol, 154: 1889, 1995 2. Kass, E. J., Majd, M. and Belman, A. B.: Comparison of pediatric renogram and the pressure perfusion study in children. J Urol, 134: 92, 1985 3. Shokeir, A. A., Provoost, A. P., El-Azab, M. et al: Renal Doppler ultrasound in children with normal upper urinary tracts: effect of fasting, hydration with normal saline, and furosemide administration. Urology, 47: 740, 1996 4. Palmer, L. S., Maizels, M., Kaplan, W. E., Firlit, C. F. and Cheng, E. Y.: Urine levels of transforming growth factor-beta 1 in children with ureteropelvic junction obstruction. Urology, 50: 5, 1997 5. Furness, P. D., Maizels, M., Han, S. W. et al: Elevated bladder urine concentration of transforming growth factor-1 correlates with upper urinary tract obstruction in children. J Urol, 162:, 1033, 1999 6. Fernbach, S. K., Maizels, M. and Conway, J. J.: Ultrasound grading of hydronephrosis: introduction of the system used by the Society for Fetal Urology. Pediatr Radiol, 23: 478, 1993 7. Seremetis, G. M. and Maizels, M.: TGF-beta mRNA expression in the renal pelvis after experimental and clinical ureteropelvic junction obstruction. J Urol, 156: 261, 1996 8. Yamamoto, T., Noble, N. A., Miller, D. E. et al: Sustained expression of TGF-beta 1 underlies development of progressive kidney fibrosis. Kidney Int, 45: 916, 1994 9. Chung, K. H. and Chevalier, R. L.: Arrested development of the neonatal kidney following chronic ureteral obstruction. J Urol, 155: 1139, 1996 10. Seseke, F., Thelen, P., Heuser, M. et al: Impaired nephrogenesis in rats with congenital obstructive uropathy. J Urol, 165: 2289, 2001 11. Isaka, Y., Tsujie, M., Ando, Y., Nakamura, H. et al: Transforming growth factor-beta 1 antisense oligodeoxynucleotides block interstitial fibrosis in unilateral ureteral obstruction. Kidney Int, 58: 1885, 2000 12. Shokeir, A. A., Provoost, A. P. and Nijman, R. J. M.: Recoverability of renal function after relief of chronic partial upper urinary tract obstruction. BJU Int, 83: 11, 1999