Pelvi-ureteric junction obstruction in children: the role of urinary transforming growth factor-β1 and epidermal growth factor

Authors


Ahmed A. Shokeir, Urology and Nephrology Center, Mansoura University, Mansoura, Egypt. e-mail: ahmedshokeir@hotmail.com

Abstract

In this section authors from Egypt describe the use of urinary TGFβ1 and epidermal growth factor in PUJ obstruction. In another paper, authors from London herald what they consider to be a new era in the management of end-stage renal disease in children, presenting the technique of bilateral synchronous posterior prone retroperitoneoscopic nephrectomy with simultaneous peritoneal dialysis.

OBJECTIVES

To investigate the role of transforming growth factor β1 (TGF-β1) and epidermal growth factor (EGF) in voided urine for the diagnosis and follow-up of children with pelvi-ureteric junction obstruction (PUJO).

PATIENTS, SUBJECTS AND METHODS

The study included 35 children with unilateral PUJO who had a pyeloplasty, and 30 healthy control children. Urine samples were obtained from the bladders of patients before surgery, and as voided samples at 1, 2, 3, 6, 9 and 12 months after surgery. Bladder urine samples were also collected from all 30 children in the control group. TGF-β1 and EGF were then measured in all the urine samples.

RESULTS

The level of bladder TGF-β1 before surgery in the patients was significantly higher than that in the healthy control group. A threshold of 190 pg/mg creatinine gave a sensitivity of 100%, a specificity of 80%, a positive predictive value of 85.4%, negative predictive value of 100% and an overall accuracy of 90.8%. Compared with the value before surgery, urinary TGF-β1 was significantly lower at 1 year after pyeloplasty. There was no significant difference between the level of EGF before surgery in the patients and that in the control group, and no significant difference in the level of EGF before and after surgery over the follow-up.

CONCLUSION

We do not recommend using EGF levels in voided urine in the routine diagnosis of children with hydronephrosis. The urinary level of TGF-β1 is a useful noninvasive tool for the long-term follow-up of children with PUJO treated by pyeloplasty. Further studies with various controls are required to confirm the diagnostic accuracy of TGF-β1 in children with PUJO.

Abbreviations
PUJO

PUJ obstruction

EGF(R)

epidermal growth factor (receptor)

ROC

receiver operating characteristic

PPV, NPV

positive, negative predictive value

UUO

unilateral ureteric obstruction.

INTRODUCTION

The response of the upper urinary tract to obstruction involves the induction of a cascade of molecular events and histological changes which involve the up-regulation of the renin-angiotensin system, with a resultant increase in the expression of tissue TGF-β1[1–3]. Honkanen et al.[4] proposed that persistently high TGF-β1 excretion correlated with morphological indices of chronicity, and the highly increased excretion suggested a persistently active and/or progressive clinical course, whereas lower values suggested a normal situation and remission. This presumes that urinary TGF-β1 reflects ongoing sclerotic and fibrotic processes in the kidneys, and that its level could be used as a noninvasive tool to assess the progression of renal disease and to follow the effects of treatments.

Increased urinary TGF-β1 excretion was reported in patients with IgA nephropathy [5], nephrotic patients with membranous nephropathy [4] and in patients with insulin-dependent or -independent diabetes mellitus [6,7]. Tsakas and Goumenos [8] reported that urinary TGF-β1 appears to be a marker of the severity of glomerular disease and of the response to treatments for inhibiting permanent damage.

PUJ obstruction (PUJO) is the most common cause of hydronephrosis in children. Currently, there is no reference standard for the diagnosis of renal obstruction in children. The diagnosis in most cases is only possible by repeated investigations, e.g. diuretic renography and Doppler ultrasonography, and by comparing changes during a longer follow-up [9]. The noninvasive measurement of biomarkers in voided urine has considerable appeal as a potential application in children with PUJO. Few studies have investigated the role of TGF-β1[10,11] or epidermal growth factor (EGF) in voided urine [12] for the clinical diagnosis of children with PUJO, and these previous studies contained few patients and had a short follow-up.

Thus we tried to overcome the deficiencies of these previous studies by providing a prospective clinical study with an adequate number of patients and controls. Moreover, we assessed the role of the two growth factors in the follow-up of these children after treatment, by measuring their values at different times over a year of follow-up after pyeloplasty.

PATIENTS, SUBJECTS AND METHODS

The study group included 28 boys and seven girls (35 renal units) with a mean (sem, range) age of 5.9 (0.71, 0.5–12) years. All patients were diagnosed with unilateral PUJO and had a pyeloplasty, and in all children the contralateral kidneys were normal. PUJO was diagnosed antenatally in eight children and the initial presentation of the remaining 27 was a flank mass in six, persistent pain in 15 and a history of UTI in six. The hydronephrosis was grade 3 or greater in all patients, according to the Society for Fetal Urology classification [13]. In all patients renography showed a half-time drainage of ≥ 20 min, and an obstructive washout curve pattern during the diuretic phase. All children had sterile urine before surgery and none had concurrent VUR. After inducing anaesthesia, urine samples were obtained from the bladder via a urethral catheter; voided urine samples were collected at 1, 2, 3, 6, 9 and 12 months after surgery. The renogram was repeated at 3 and 12 months after surgery.

The study group was compared with a control group, comprising 30 age-matched healthy children (16 boys and 14 girls, mean age 6.17 years, sem 0.51, range 0.75–13). Bladder urine samples were collected from all 30 children in the control group.

The TGF-β1 concentration in the urine was determined in duplicate by a quantitative sandwich ELISA kit (Emax Immunoassay System, ♯G7590, Promega Corporation, Madison, WI, USA). The EGF concentration in the urine was determined in duplicate by an ELISA kit (Quantikine; R and D Systems, ♯DEG00; Minneapolis, MI, USA). Urinary creatinine was determined by a modified-rate Jaffe’s technique [14] on a Synchorn CX7 system (Beckman Instruments, Inc., USA). The local ethical committee approved the study

The results were assessed statistically using a t-test and simple linear regression analysis. The receiver operating characteristic (ROC) curve was constructed to determine the threshold values of the two biomarkers giving the best sensitivity and specificity.

RESULTS

Before surgery, the level of urinary TGF-β1 in the patients was significantly higher than that in the healthy control group, at 374 and 157 pg/mg creatinine, respectively, giving a ratio of 2.4 : 1. The level of TGF-β1 was unaffected by the clinical presentation. After surgery, TGF-β1 showed an initial significant increase in the first month and started to decrease gradually until the level became significantly lower at 1 year than the baseline value (Fig. 1a). The level of urinary TGF-β1 in the patients at 1 year after pyeloplasty was not significantly different from that in the control group, at 186 (9) and 157 (13) pg/mg creatinine, respectively (P = 0.07).

Figure 1.

The mean (sem) TGF-β1 and EGF levels in the patients and controls: P1, values in patients vs values in the control group; P2, values before vs after surgery in patients.

In the patient group, eight children were ≤ 1 year old and the remaining 27 were >1 year old. The younger children had significantly higher levels of urinary TGF-β1, at 601 (98) pg/mg creatinine, than the older ones, at 307 (36) pg/mg creatinine (P < 0.01).

Using the ROC curve, the threshold urinary TGF-β1 level of the patients and controls which gave the highest sensitivity and specificity was 190 pg/mg creatinine; this threshold gave a sensitivity of 100%, a specificity of 80%, a positive predictive value (PPV) of 85.4%, a negative PV (NPV) of 100% and an overall accuracy of 90.8% (Table 1). For the eight children aged ≤ 1 year a threshold of 330 pg/mg creatinine gave a sensitivity of 100% and a specificity of 100% (Table 1). For the 27 children aged ≥ 1 year, a threshold of 200 pg/mg creatinine gave a sensitivity of 100%, a specificity of 76.7%, a PPV of 79.4%, a NPV of 100% and an overall accuracy of 87.7% (Table 1).

Table 1. 
The distribution of urinary TGF-β1 and EGF in children with PUJO
Growth factor (reference value)Reference standard (renography)
(+ve)(−ve)
Number of patients
TGF-β1, pg/mg
All patients (190 pg/mg creatinine)
 +ve35 6
 −ve 024
≤1 year (330 pg/mg creatinine)
 +ve 8 0
 −ve 030
>1 year (200 pg/mg creatinine)
 +ve27 7
 −ve 023
EFG, ng/mg
All patients (40 ng/mg creatinine)
 +ve14 6
 −ve2124
>1 year old (50 ng/mg creatinine)
 +ve2214
 −ve 516

The split renal function of the corresponding kidney before, 3 months, and after 1 year after pyeloplasty was 27.9 (1.08)%, 32.9 (0.75)% and 34.2 (1.08)%, respectively, showing a significant and progressive improvement (P < 0.001 for each). A possible influence of split renal function on the level of urinary TGF-β1 was also evaluated; there was no significant linear correlation of TGF-β1 with the split renographic clearance in the patient before surgery and during the follow-up.

For EGF, there was no significant difference between the level in the patients, at 54.5 (4.77) ng/mg creatinine, and that in the control group, at 54.3 (4.35) ng/mg creatinine. Compared to the baseline value of the patients there was no significant difference in the level of urinary EGF after surgery and during the follow-up (Fig. 1b). The level of urinary EGF during the 1 year of follow-up was not significantly different from the level in the controls.

The patients aged ≤ 1 year had significantly higher levels of urinary EGF before surgery, at 91.0 (5.04) ng/mg creatinine, than the older patients, at 43.7 (4.11) ng/mg creatinine. The threshold for urinary EGF of the patients and controls that gave the highest sensitivity and specificity was 40 ng/mg creatinine, giving a sensitivity of 40%, a specificity of 80%, a PPV of 70%, a NPV of 53.3% and an overall accuracy of 58.5% (Table 1). We also used the ROC curve for the two age groups (≤1 and >1 year). The younger group gave an irregular curve with no reference value, but a threshold of 50 ng/mg creatinine for the older group gave a sensitivity of 81.5%, a specificity of 53.3%, a PPV of 61.1%, a NPV of 76.2% and an overall accuracy of 66.7% (Table 1).

DISCUSSION

TGF-β1 is the main modulator of the healing process after tissue injury; normally, its release ceases by feedback mechanisms when the healing process has been completed, but if TGF-β1 release is not switched off, extracellular matrix components are accumulated and tissue fibrosis occurs [15]. Up-regulation of TGF-β1 synthesis in the kidney is followed by accumulation of collagen and scarring [16]; it could participate as a key factor in the common mechanisms leading to tissue fibrosis and the development of advanced chronic renal disease of various causes [17], whereas the administration of specific antiserum against TGF-β1 results in amelioration of renal damage [8].

There is agreement between the present data and those of Furness et al.[10] and Liatsikos et al.[18], who reported that bladder urinary TGF-β1 in children with upper urinary tract obstruction was significantly higher, by up to four times, than that in controls. Thus bladder TGF-β1 could be used clinically to confirm the diagnosis of PUJO in children.

Furness et al.[10] showed that the mean bladder urinary TGF-β1 level was lower in older children than in those aged <2 years. Similarly, the present mean urinary TGF-β1 level in the patients was lower in older children than in those aged ≤ 1 year. This lower urinary TGF-β1 level in older children with obstruction might reflect a lower and more steady-state production of this growth factor with long-standing obstruction [11]. In younger children the higher level of urinary TGF-β1 increased its diagnostic accuracy to 100%. Nevertheless, because there were few young children in the present study, this diagnostic accuracy should be confirmed by similar studies using the same methods in more patients. Moreover, the higher level of TGF-β1 in younger children might be attributable to the lower creatinine production in early childhood, rather than a true increase in TGF-β1 production.

The correlation between urinary TGF-β1 levels before and after surgery for PUJO was first evaluated in a clinical setting by El-Sherbiny et al.[11], but that study had a limited follow-up of 3 months. The authors showed that the mean bladder TGF-β1 level 3 months after surgery tended to decrease, albeit still not significantly different from the level before surgery. The present data showed a transient initial increase in urinary TGF-β1 level during the first month after surgery, with a persistent gradual decrease thereafter. The decrease became statistically significant at 1 year of follow-up, indicating that the urinary level of TGF-β1 could be used as a noninvasive tool in the long-term follow-up after pyeloplasty of children with PUJO. The initial increase might be attributable to the requirement of this growth factor in successful healing of the PUJ [18].

Notably, there was a difference in the thresholds between the present and similar previous studies [10,11]. This difference is attributable to the difference in the method of measuring TGF-β1, but the ratio between the level of TGF-β1 in the patients and controls were similar in the present series and the other studies [10,11]. Our results are in agreement with these earlier findings [10,11], where there was no linear correlation between split renographic clearance and urinary TGF-β1 levels in children with upper urinary tract obstruction.

One of the limitations of the present study is that the increase in the level of TGF-β1 is not specific to obstructive uropathy. It can increase in other types of nephropathy, e.g. nephrotic syndrome [19], glomerulosclerosis [4], IgA nephropathy [5] and diabetes [6,7]. Other studies included a wider range of control groups are invited to evaluate the role of TGF-β1 in the diagnosis of PUJO. Nevertheless, TGF-β1 is useful as a noninvasive tool in the long-term follow-up of children with PUJO.

EGF is a peptide that stimulates proliferation and acts as a survival factor in the developing rat kidney [20]. EGF is normally synthesized by distal tubular cells, with increasing expression during maturation [21]. Chronic unilateral ureteric obstruction (UUO) suppresses renal EGF production in neonatal rats or in infants and children [12,21]. Exogenous EGF reduces tubular apoptosis by 80% in the neonatal rat with chronic UUO, and enhances recovery after the relief of obstruction. Exogenous EGF also inhibits tubular apoptosis in the adult rat subjected to UUO, while in neonatal wild-type mice, exogenous EGF promotes apoptosis instead of cell survival in the obstructed kidney [22].

The maturation and proliferation of kidney cells occurs through the potential role of the EGF receptor (EGFR) and its ligand (EGF) in cell division. Lin et al.[23] stated that EGFR and its ligand might function together as a transactivation complex, and this can bind to specific DNA sequences to activate the gene expression required for highly proliferative activities. Thus reductions in EGF levels might reflect reduced EGFR signalling.

Grandaliano et al.[12] reported significantly less urinary EGF in a group with PUJO than in controls, a finding not supported by the present study, although we used a similar technique. This disagreement is probably due to the difference in the source of urine samples between the studies; we used bladder urine samples in all the patients, while Grandaliano et al.[12] used renal pelvic samples in 10 and voided urine samples in 14 of their patients.

We showed that urinary EGF levels in children with PUJO are of no value in children aged <1 year and are of limited clinical value in older children. Based on these results we do not recommend using urinary EGF levels in the routine diagnosis of children with hydronephrosis. Urinary TGF-β1 is a useful noninvasive tool for the long-term follow-up of children with PUJO after pyeloplasty. Further studies with various controls are required to confirm the diagnostic accuracy of TGF-β1 in children with PUJO.

CONFLICT OF INTEREST

None declared.

Ancillary