The change in renal function in the supranormal hydronephrotic kidney after pyeloplasty

Authors

  • Cheryn Song,

    1. Department of Urology, University of Ulsan College of Medicine, Asan Medical Center, Seoul, and Ulsan University Hospital, Ulsan, Korea
    Search for more papers by this author
  • Hongjoo Park,

    1. Department of Urology, University of Ulsan College of Medicine, Asan Medical Center, Seoul, and Ulsan University Hospital, Ulsan, Korea
    Search for more papers by this author
  • Sungchan Park,

    1. Department of Urology, University of Ulsan College of Medicine, Asan Medical Center, Seoul, and Ulsan University Hospital, Ulsan, Korea
    Search for more papers by this author
  • Kyung Hyun Moon,

    1. Department of Urology, University of Ulsan College of Medicine, Asan Medical Center, Seoul, and Ulsan University Hospital, Ulsan, Korea
    Search for more papers by this author
  • Kun Suk Kim

    1. Department of Urology, University of Ulsan College of Medicine, Asan Medical Center, Seoul, and Ulsan University Hospital, Ulsan, Korea
    Search for more papers by this author

Kun Suk Kim, Department of Urology, University of Ulsan, College of Ulsan Asan Medical Centre, 388–1 Poongnap-dong, Songpa-gu, Seoul, Korea (138–736).
e-mail: kskim2@amc.seoul.kr

Abstract

OBJECTIVE

To investigate changes in the differential renal function (DRF) before and after pyeloplasty in renal units with unilateral pelvi-ureteric junction obstruction (PUJO) with supranormal function, and to evaluate the clinical significance.

PATIENTS AND METHODS

We reviewed the medical and radiographic records of 29 children (26 boys and three girls) with unilateral PUJO with a DRF (estimated by 99mTc-mercaptoacetyltriglycine renal scintigraphy) of ≥ 50% in the affected renal unit, who had pyeloplasty and were followed for >1 year after surgery. Patients were divided into two groups according to the degree of change in their DRF to compare the clinical variables, anteroposterior pelvic diameter and parenchymal thickness measured by renal ultrasonography. The mean (range) follow-up after pyeloplasty was 35 (12–89) months.

RESULTS

After pyeloplasty, although the mean DRF reduced from 53.8% to 51.4%, in seven (24%) patients the DRF decreased significantly (>5%) while most (76%) showed a change of ≤ 5% of the preoperative function, or further improvement. In the seven patients with a DRF of ≥ 55% before surgery, the DRF afterward was >55% in four and 50–55% in one. Between those with and with no significant reduction in DRF, only the preoperative renal parenchymal thickness differed significantly (2.78 vs 5.00 mm, P = 0.006).

CONCLUSION

Supranormal DRF exists and represents the true split function of the affected renal unit in patients with adequate renal parenchyma; these units maintained the supranormal function after pyeloplasty. Parenchymal thickness might be useful to differentiate between the true and false estimates and predict surgical outcome before surgery.

Abbreviations
DRF

differential renal function

APPD

anteroposterior pelvic diameter

PT

parenchymal thickness

RUS

renal ultrasonography.

INTRODUCTION

Differential renal function (DRF), as obtained from a radionuclide renogram, has become the most important criterion in therapeutic decision-making in children with PUJ obstruction (PUJO). However, in the presence of a normally appearing contralateral kidney, a DRF of 55% or ≥ 50% measured from the affected renal unit has been reported to constitute 9–21.7% of these patients, depending on the definition [1,2]. Such a paradoxical phenomenon has caused controversy about its existence and the clinical or prognostic significance [2–5].

Previously we reported that a supranormal measured DRF is not an artefact and suggested its existence using various methods of background subtraction for the calculation of radioisotopic estimates [6]. Recent reports agree that renal function remains relatively unchanged after pyeloplasty, especially in PUJO detected in utero[7–10]. Thus, if a supranormal DRF was a quantification error resulting from the reservoir effect of the radioisotopes, the DRF after surgery would be dramatically reduced in patients in whom urinary obstruction was relieved. Alternatively, if it is exists, the DRF after surgery would remain supranormal in most patients. In the present study we report the follow-up results after pyeloplasty, to verify the existence of a supranormal DRF, confirming our previous study, and to clarify its clinical significance.

PATIENTS AND METHODS

From 1995 to 2004, 112 children had a dismembered pyeloplasty by one surgeon at the Asan Medical Centre. Of these patients, 39 with no adequate preoperative evaluation records or sufficient follow-up data (>1 year after surgery), eight with concomitant congenital anomalies and 17 with bilateral PUJO were excluded. Of the remaining 48 children, 29 (26 boys and three girls) had a DRF of ≥ 50% on their initial 99mTc-MAG3 renal scan, and comprised the final study group. The DRF from the renal scan was obtained according to our internally standardized protocol [11]. With no bladder catheterization, unless they had evidence of VUR or bladder dysfunction, patients were hydrated i.v. at 10 mL/kg with normal saline over 30 min before injection with 99mTc-MAG3 (3.7 MBq/kg) after which one experienced technologist calculated the DRF and generated renograms. To measure DRF, regions of interest were drawn over the entire kidney and the background on each side. The kidney background was manually drawn on 1–2-min images in a crescent shape over the outer aspect of the kidney, to exclude the collecting system. The DRF was then calculated as the percentage of relative activity of the hydronephrotic kidney over the sum of background-corrected total renal activity at 1–2 min after the injection. We used this perirenal background correction method and measurement of DRF at 1–2 min after injection to avoid potential overestimation.

After surgery, patients were followed with renal ultrasonography (RUS) and 99mTc-MAG3 renal scintigraphy, beginning at 3 and 12 months, then annually if there were no further complications. From the RUS the anteroposterior pelvic diameter (APPD) was measured from the axial image at the hilar level, and the parenchymal thickness (PT) and kidney size were measured from the coronal image in the midplane. From the diuretic renogram, the presence or absence of obstructive drainage was evaluated in addition to the DRF of the affected renal unit. The mean (range) follow-up was 35 (12–89) months, and although all the available results were reviewed, the results at 1 year after surgery were used for the comparative analysis.

To analyse and compare the variables affecting renal function and surgical outcome, patients were divided into two groups according to the degree of change in their DRF, i.e. those with or with no significant reduction in DRF (reduced by ≥ 5% vs <5% or increase by ≥ 5%). In both groups, the APPD, PT, kidney sizes and the ratio of the affected to contralateral kidney size, measured by RUS before and after surgery, were compared, as were the clinical characteristics. We used nonparametric tests (Wilcoxon rank-sum test to compare ultrasonographic variables, and Fisher’s exact test for patient characteristics), with P ≤ 0.05 taken to indicate significant differences.

RESULTS

In all, 29 children (mean age 1.7 years, range 1 month to 10 years) with a initial DRF of ≥ 50% had pyeloplasty for unilateral PUJO. Of these patients, 22 had a DRF of <55% while seven had a DRF of ≥ 55%. The patients’ characteristics before surgery, other than the initial DRF, including mode of presentation or history of symptomatic UTI, did not differ between the groups (Table 1). Furthermore, there were no diagnostic abnormalities found in the unaffected contralateral renal units on RUS. All patients had an obstructive pattern (T1/2 >20 min) on the preoperative radionuclide scans.

Table 1. 
The preoperative characteristics of patients according to the initial DRF
CharacteristicDRF ≥50%≥55%
Number29 7
Mean age, years 1.7 1.1
Gender (M:F)26 : 3 7 : 0
N (%) or n
 Prenatal detection24 (83) 7
 History of symptomatic UTI 6 (21) 1
Mean dimension, mm (by RUS)
 APPD31.130.6
 PT 4.4 3.8
DRF, %53.859.4
Follow-up, months35.044.2

Follow-up RUS at 1 year showed an improvement in hydronephrosis and compensatory kidney growth in both groups. The mean APPD decreased from 31.1 to 12.6 mm (mean decrease 18.2) and the mean PT increased from 4.4 to 10.1 mm (mean increase 5.6, both P < 0.001). There was a similar improvement in patients with an initial DRF of ≥ 55% (APPD, 30.6 to 14.0 mm, mean decrease 13.0 mm; PT, 3.8 to 8.1 mm, mean increase 3.7, both P < 0.001). On radionuclide scans, although the mean DRF decreased from 53.8% to 51.4% (mean 2.4%, P = 0.028), in 76% of the patients the difference was <5% (Fig. 1).

Figure 1.

Changes in DRF in patients with an initial DRF of ≥ 50%.

The seven patients with an initial DRF of ≥ 55% had similar characteristics before surgery (Table 2). All patients had presented with prenatally detected PUJO and were asymptomatic during the follow-up. After surgery, except for the two patients in whom the DRF reduced to 38% and 44% (numbers 3 and 5, respectively), the DRF remained within the same preoperative, supranormal range. An increase (numbers 1 and 2) or decrease in DRF (numbers 4 and 7) were mostly <5%, except for one patient (number 6) in whom the DRF decreased by 13% but remained in the supranormal range.

Table 2.  Clinical characteristics and surgical outcome of individual patients with an initial DRF of ≥ 55%. All patients had hydronephrosis grade 4 (Society of Fetal Urology system) and all were diagnosed prenatally
VariablePatient
1234567
Age, years 4 0.5 0.08 0.17 0.17 0.58 2
Before surgery
APPD, mm402522.620.24038.827.6
PT, mm 4.0 7.0 3.0 3.0 3.0 1.5 5.0
DRF, %58.55655.460557556.2
After surgery
APPD, mm2013 9142712.714
PT, mm 5.012.0 8.0 8.0 8.6 6.58 7.5
DRF, %65603858446253.4

Of the 29 patients, the DRF improved or reduced by ≤ 5% in 22 (76%) while in seven (24%) it was reduced by more than 5%. Between those with a significant reduction in DRF and those in whom the DRF remained relatively unchanged or even improved, there were no significantly different clinical characteristics (Table 3). However, the parenchyma before surgery was significantly thinner in patients with a significant reduction in DRF (2.78 vs 5.00 mm, P = 0.006). Other ultrasonographic variables, including the kidney size or the kidney-to-kidney ratio, showed no significant relationship with the degree of change in DRF after surgery.

Table 3. 
A comparison of variables before surgery in patients with and with no significant reduction in DRF after surgery
VariableDRFP
reduced by >5%change ≤ 5% or increase by >5%
  • *

    ratio of affected to contralateral kidney size.

N (%) or n:
Total 7 (24)22 (76) 
Mean age, years 1.9 1.60.438
Male 719 (86)0.557
Prenatal presentation 519 (86)0.767
Documented UTI 0 6 (27)0.289
Mean dimension (by RUS), mm
 APPD33.6430.450.303
 Parenchymal thickness 2.78 5.000.006
 Kidney size 8.57 8.430.767
 Kidney size ratio* 1.39 1.390.668

DISCUSSION

A supranormal measured DRF in renal units affected by PUJO, regardless of its definition, can be perplexing. When the affected kidney is severely hydronephrotic, with thinned parenchyma, while the contralateral renal unit appears normal with an undilated pelvicalyceal system and adequate PT, the DRF does not seem to be a valid estimate. Accordingly, controversy about the existence of a supranormal kidney in congenital hydronephrosis has persisted [1–5]. Previously, we reported evidence supporting the existence of the supranormal kidney; these renal units had a greater than normal DRF in earlier phases of the radionuclide scan, before the pelvic retention of the radio-isotopes began, and left kidneys in which the hepatic backgrounds were excluded consistently had a high DRF, and with all background subtraction methods [6]. However, in our previous study we could not interpret the clinical or prognostic significance. In the present study, at a mean follow-up of 35 months after pyeloplasty, the DRF remained relatively unchanged in 22 (76%) of 29 children, with 16 (73%) of these within their initial range, confirming that supranormal DRF exists in some patients. Of the seven patients who had a significant reduction in DRF after pyeloplasty it is possible that the supranormal DRF was erroneously overestimated. The two groups could be distinguished, from the present results, by the preoperative PT measured on RUS. In patients with adequate parenchyma before surgery there was a supranormal DRF, correctly measured, implying that the affected renal unit is the better functioning of the two, and thus might be expected to remain supranormal after surgery. Conversely, in those with insufficient parenchyma, the initial DRF might have been overestimated and that after surgery anticipated to decrease significantly. The cause of the supranormal function in unilateral PUJO was hypothesized to be single-nephron hyperfiltration [1] or an effect of larger kidneys, with no significant reduction in glomerular units [12], until Ham et al.[13] reported no significant difference in the mean glomerular areas between the hydronephrotic kidneys and the normal controls. From the results of their study, they advocated rejecting the theory of increased nephron volume as a cause of supranormal DRF. According to the present results, PT but not kidney size differed between those with and with no significant reduction in DRF, supporting the view that supranormal function is due to more rather than a greater volume of glomerular units. Previously, the process of compensatory hypertrophy was shown to results from an increase in glomerular number and volume. Studies showed more cellular components of the glomeruli with notable capillary elongation and duplication [14], or an increase in glomerular RNA content where no morphometric enlargement was noted [15], suggesting true glomerular hypertrophy. However, single-nephron hyperfiltration might explain the overestimated DRF. In response to increased hydraulic pressure, increases in the capillary component of the glomeruli, if not the entire glomerular unit, were reported, which would result in increased blood flow and the resultant hyperfiltration leading to transient hyperfunctioning of these renal units. With no accompanying morphological adaptation, such transient hyperfunction would fade after the relief of obstruction, as was shown by the significant reduction in DRF in the present seven patients.

Changes in DRF after pyeloplasty, despite compensatory growth in these renal units [1,16], will vary, and recent reports agree that pyeloplasty improves the DRF only in patients with postnatal symptomatic PUJO, whereas prenatally diagnosed asymptomatic patients are almost unaffected [9,10,17]. In a study by Calisti et al.[10], of the 17 patients with antenatally detected PUJO who had an early pyeloplasty (before a recorded deterioration in DRF), 11 (65%) showed limited functional improvement (not statistically significant). In their study, delayed pyeloplasty (after recording a deterioration in DRF) failed to restore the initial DRF in 72% of the patients. Similarly, in the present study the increase or decrease in DRF after pyeloplasty in most patients was minimal and was not statistically significant. In the seven patients with a significant reduction in DRF, the decrease was noted immediately after surgery and remained, thus it might be assumed that the DRF before surgery was overestimated, confirming that true renal function remains relatively unchanged by pyeloplasty.

However, in previous studies, no clinical or histological variables predictive of the improvement or deterioration of DRF after surgery were determined, and many investigators recommended early surgery, as such discrimination was difficult [7–10]. The present results suggest that in the subgroup of patients who have a DRF of ≥ 50%, the PT measured by RUS can be used before surgery to differentiate whether the estimates are correct or overestimated, and to predict the surgical outcome. Accordingly, patients with adequate renal parenchyma in the affected renal unit indicate that the functionally dominant unit is affected and should require timely intervention. Alternatively, patients with no adequate renal parenchyma, a supranormal DRF might have been overestimated, implying that prompt surgical correction is necessary to prevent further functional loss in the obstructed unit.

In conclusion, a supranormal DRF in unilateral PUJO can be real and represents the true split function of the affected renal unit in patients with adequate renal parenchyma; these units can be expected to maintain supranormal function after pyeloplasty. Parenchymal thickness measured by RUS might be useful to differentiate between true and false estimates of DRF and to predict the surgical outcome.

CONFLICT OF INTEREST

None declared.

Ancillary