Outcome Analysis of Reduction and Nonreduction Dismembered Pyeloplasty in Ureteropelvic Junction Obstruction: A Randomized, Prospective, Comparative Study.

Objectives: The objective of this study is to compare the changes in renal function and drainage following open dismembered pyeloplasty with and without renal pelvis reduction. Materials and
Methods: Randomized prospective study of children with ureteropelvic junction obstruction undergoing pyeloplasty with (Group 1) and without (Group 2) pelvis reduction over an 18-month period. Postoperative function and drainage were assessed by ethylene dicysteine (EC) scan and intravenous urography (IVU) and renal pelvis size by ultrasonography.
Results: Forty-two patients (2 months-11 years) participated. The mean preoperative EC scan function was Group 1: 45.88% ± 14.42% (5%-80%) and Group 2: 39.22% ± 9.75% (21%-53%). (P = 0.117). The mean postoperative EC scan function of Group 1 was 42.64% ± 9.62% (17%-54%) and 43.75% ± 9.88% (17%-58%) and of Group 2 was 44.77% ± 12.82% (20%-68%) and 42.25% ± 8.56% (23%-58%) at 3 months (P = 0.584) and ≥ 1year (P = 0.385), respectively, with no significant difference. None required re-do pyeloplasty. The number of patients with slow drainage, especially at 3 months and also at ≥1-year postoperative period on EC scan was slightly higher in Group 2 compared to Group 1 but did not attain statistical significance. There was postoperative improvement in function and drainage on IVU with no significant difference between the two groups, (P = 0.214; P = 0.99, respectively). At a mean follow-up of 45.5 months, Group 2 also showed significant reduction in pelvis size on ultrasound (P = 0.011).
Conclusion: Postoperative function remained stable in both groups. More number of patients achieved unobstructed drainage by 3 mo postoperative after reduction pyeloplasty but drainage patterns were mostly similar between reduction and nonreduction of pelvis group in late follow-up. Copyright:

INTRODUCTION

Ureteropelvic junction obstruction (UPJO) is thought to be the cause of about 10%–30% of all cases of congenitally significant hydronephrosis, with an incidence of 1 in 1250 births.[123]

Dismembered pyeloplasty was described in 1949 by Anderson and Hynes and became the procedure of choice for UPJO, with excellent long-term results.[4] It involves complete removal of the narrowed dysfunctional segment, tailoring of the renal pelvis if necessary, and re-approximation of the ureter to the renal pelvis in a dependent position.[5] Excision of the renal pelvis can prevent urine stasis behind a newly created anastomosis and avoid ureteral kinking.[6] However, unnecessary surgery might lead to problems such as urine leakage or long hospitalisation.[6] With the increasing popularity of laparoscopic dismembered pyeloplasty, the need for reduction of the renal pelvis is being re-evaluated.[78910] Many also believe that the redundant renal pelvis is protective, and removing it may deprive the kidney from its benefit at a time when it may need it the most. The controversy, hence, continues. Many studies have shown that in the majority of cases, renal function remains stable after pyeloplasty.[11] Limited studies are available concerning the effects of pelvic reduction on outcome.[6912] In this study, we compared the postoperative renal function and drainage with or without reduction of pelvis during open dismembered pyeloplasty.

MATERIALS AND METHODS

This randomized prospective study was conducted in a tertiary care pediatric surgery referral centre. All UPJO patients between 1 month and 12 years age of either sex who underwent pyeloplasty between July 2013 and December 2014 were included. Patients with ureteral dilatation, hydronephrosis in duplex system, ipsilateral or contralateral reflux, fused kidney, anatomical or neuropathic abnormality of the lower urinary tract and redo pyeloplasty were excluded. Approval was obtained from the Institute Ethics Committee prior to the study.

Preoperative evaluation included complete hemogram, blood urea, serum creatinine, and urine culture and sensitivity. In the ultrasound kidney ureter bladder (USG KUB) the following were noted: renal pelvis anteroposterior diameter (APD) (≥2 cm, or 1–2 cm with other parameters mentioned below), abrupt cut off at the ureteropelvic junction (UPJ), nonvisualization of ureter, calyceal separation (mild: splaying, moderate: ballooning of major and minor calyces, severe: massive dilatation of pelvis and calyces), renal parenchymal thickness (normal: [As per nomograms for age and side of kidney], mild thinning: below normal range, [usually between 6 and 9 mm] and significant thinning [<5 mm]).

The decision to operate was not based solely on USG KUB as a dilated pelvis can be due to other causes. In intrarenal pelvis especially in crossing vessels, the APD may not be significant but other parameters may be abnormal. In conservatively managed hydronephrosis, an increase in dimensions warrant further investigations.

All patients underwent ethylene dicysteine (EC) renal dynamic scan with frusemide being injected intravenously along with the tracer (F 0 protocol) to avoid the disturbance of a second injection during the 25-min dynamic acquisition. Different terminologies used in drainage assessment are given below with composite scans for the same being shown in Figure 1.

Figure 1

Representative scans of slow but unobstructed drainage, unobstructed drainage, obstructed drainage and slow / delayed drainage on ethylene dicysteine scan of study patients

Grade 3: Unobstructed drainage (UOD): Complete clearance of all tracer from the Pelvicalyceal system (PCS) during the 25-min dynamic study with a normal renogram curve pattern. The pre- and post-void static images show no residual tracer in the PCS. A 3-h static image is therefore not required.

Grade 2: Slow UOD (SUOD): Incomplete clearance from the PCS is seen during the initial 25-minute dynamic study, with moderate to significant residual tracer in the PCS in the immediate postdynamic pre- and post-void static images followed by complete clearance from the PCS by 3 hours (the 3-h static image shows no residual tracer in the PCS). Occasionally, when there is only partial clearance by 3 h, a 5-h static image shows complete clearance from the PCS. The time required may vary, depending on the kidney. The renogram curve (representing the drainage pattern during 25-min dynamic study) is usually a rising pattern, or may show plateauing or mild down-sloping after an initial rise.

Grade 1: slow or delayed drainage: Tracer clearance from the PCS is equivocal. The initial 25-min dynamic study shows partial or incomplete tracer clearance from the PCS, with some residual tracer in the PCS (which may not be significantly dilated) in the pre- and post-void as well as 3-h delayed static images. A 5-h static image (when acquired) may show mild residual tracer remaining in the PCS. Often seen in hydronephrotic kidneys with moderate to severely impaired cortical function (causing slow tracer accumulation). “Delayed” signifies a slower drainage than “slow,” and requires close follow-up. Repeat study with F-15 protocol may be considered. A stable function in an asymptomatic patient, does not warrant redo surgery.

Grade 0: Obstructed drainage (OD)

We use a combination of the following criteria to identify UPJO on EC scan:

A continuously rising renogram pattern. This is not confirmatory of UPJO by itself, as it may be seen in hydronephrotic kidneys, an inadequately hydrated patient or extravasated tracer injection

No clearance of the tracer during the initial 25-min dynamic study, with subsequent retention of tracer in the PCS in the pre- and post-void static images upto 3 hours after administration of the tracer with intravenous frusemide. In case of any doubtful excretion, or in compromised renal cortical function with slow accumulation of tracer in the kidney, a 5-h delayed image is taken in specific patients to confirm the findings of the 3-h static image.

EC scan drainage was assessed in all patients, considering each kidney in bilateral UPJO cases as separate units. EC scan function comparison was done only for unilateral UPJOs and was graded as: (a) >5% reduction of function (deterioration), (b) ± 5% change of function (stable) and (c) >5% increase in function (improvement).[121314]

Intravenous urography (IVU) was also done. We do both EC scan and IVU to be doubly sure about the diagnosis. While EC scan is superior in giving renal function in the presence of a nonvisible/poorly visible kidney, IVU gives structural information about the kidney and ureter especially in duplex systems. EC scan may wrongly diagnose cases of partial or complete vesico ureteric junction obstruction as UPJO or wrongly give a higher function in associated vesicoureteral reflux or slow draining systems. It is also our experience that the improvement in function and drainage is better appreciated by IVU comparison following pyeloplasty in a severely obstructed dilated system. Moreover, in IVU, each kidney can be studied as a separate unit even in bilateral cases or solitary kidney.[15]

For IVU drainage, grading was as follows:

Grade 4: Absence of contrast by 2 h film or 5 min post Lasix injection

Grade 3: Absence of contrast after 10 min of Lasix injection

Grade 2: Absence of contrast after 20 min of Lasix injection

Grade 1: Presence of contrast even after 20 min of Lasix injection and

Grade 0: Nonvisualized kidney.

The persistence of the contrast in the PCS 2 hours after its injection and its further retention even 20 min after lasix is injected was the criteria used to diagnose UPJO.

The uptake of contrast at different time periods was noted to assess the IVU function and graded as follows:

Grade 7: If uptake is prompt within 7 min which is the first study

Grade 6: If first uptake is at 15 min

Grade 5: If first uptake is at 30 min

Grade 4: If first uptake is at 1 h

Grade 3: If first uptake is at 2 h

Grade 2: If first uptake is at 2–4 h or post Lasix 5 and 10 min

Grade 1: If first uptake is at >4 h or post Lasix 20 min and

Grade 0: Nonvisualized kidney.

Informed consent was taken from parents regarding the surgery including the possibility of reduction or nonreduction of pelvis. Patients were randomized by envelope method on the morning of the surgery into two groups, namely: Group 1: reduction of pelvis and Group 2: nonreduction of pelvis.

At induction of anesthesia, all patients were administered similar antibiotic intravenously (injection cefotaxime 50 mg/kg) which was continued postoperatively for 48 h twice daily. Patients were operated by either one of two surgeons. Open modified Anderson-Hynes pyeloplasty was performed through an anterolateral extra peritoneal flank incision. The ureter was laid open along its lateral margin in a downward direction from the UPJ site and the entire narrow segment excised till normal ureter was opened up for 2 cm followed by a wide pelvi-ureteric anastomosis. In Group 1, the pelvis reduction was done, leaving only 1 cm tissue from the renal sinus. In Group 2, the renal pelvis was not excised.

Appropriate size Double J (DJ) stent was placed in all the patients. Patients were discharged on the 2nd or 3rd postoperative day if oral intake was adequate and there were no complications. DJ stents were removed on outpatient basis under short general anesthesia, 2 weeks after surgery.

Postoperatively, patients were followed up with: EC scan at 3 months (EC1) and ≥1 year (EC2) and IVU once at 3 months after surgery. The change in renal drainage and differential renal function (DRF) between the two groups before and after surgery were compared. Pelvis dimensions on USG KUB taken at least 1 year after pyeloplasty were used for comparison between the 2 groups to look for similarities and dissimilarities with time. All patients were continuously followed up for 5 years after surgery.

Postoperative complications including urinary tract infection were noted and treated accordingly.

Statistical analysis: Statistical Package for Social Sciences (SPSS Inc., Chicago, IL, USA version 17.0 for Windows) was used. Comparison of quantitative variables between the study groups was done using the Mann–Whitney U-test for independent samples. For comparing categorical data, the Pearson Chi-square test was performed. A probability P < 0.05 was considered statistically significant.

RESULTS

Forty-two patients fulfilled the inclusion criteria and underwent pyeloplasty during the study period from July 2013 to December 2014. The patients' age ranged from 2 months to 11 years with mean of 2.75 ± 2.7 years in Group 1 (n = 21) and 2.98 ± 4.21 years in Group 2 (n = 21) (P = 0.84). There were 3 females (14.3%) and 18 (85.7%) males in both the groups. In Group 1, the left kidney was involved in 10 (47.6%) patients, the right kidney in 8 (38.1%) patients and both kidneys in 3 (14.3%) patients. In Group 2, the left kidney was involved in 14 (66.7%) patients, right kidney in 5 (23.8%) patients and 2 (9.5%) patients had bilateral UPJOs. There was one solitary kidney with UPJO in each group. Hence, there were a total of 24 renal units in Group 1 and 23 renal units in Group 2.

In Group 1, 8 (38.1%) patients and in Group 2, 13 (61.9%) patients were antenatally diagnosed with hydronephrosis (P = 0.123). Both the groups had similar presentation and normal routine hematological and biochemical investigations.

Preoperative APD of the renal pelvis in the transverse plane on USG KUB showed a mean of 3.14 ± 2.28 cm (range 0.7–11 cm) in Group 1 and 2.38 ± 1.36 (range 0.9–6 cm) in Group 2 (P = 0.4795). Preoperative lower pole calyceal separation showed a mean 7.65 ± 2.43 mm (5-13 mm) in Group 1 and 8.29 ± 3.73 mm (4–16 mm) in Group 2. Parenchymal thickness showed a mean 5.3 ± 1.76 mm (2.7–10 mm) in Group 1 and 6.34 ± 2.5 mm (3.4–11.7 mm) in Group 2. One patient in Group 2 had a small kidney with 9 mm pelvis but OD on EC scan and IVU. Five patients in Group 1 and two patients in Group 2 had APD above 5 cm. There was a history of preoperative percutaneous nephrostomy tube placement for short periods of time before surgery for grossly enlarged pelvis with thin parenchyma or pyonephrosis in 4 patients in Group 1.

Intraoperatively, there was intrinsic narrowing at the UPJ in 45 renal units, of whom 24 were in Group 1 and 21 in Group 2. Two patients in Group 2 had aberrant crossing vessel with no intrinsic obstruction.

For comparison of DRF on EC scan, patients with solitary kidney (Group 1, n = 1; Group 2, n = 1) and bilateral UPJO (Group 1, n = 3; Group 2, n = 2) were excluded. Preoperative DRF was therefore compared between 17 units in Group 1 and 18 units in Group 2 and was not statistically different (P = 0.117) [Table 1]. There was no significant difference between the two groups in the IVU function preoperatively (P = 0.448). Preoperatively, the 2 groups were comparable for IVU drainage (P = 0.579) with 18 units (75%) in Group 1 and 15 units (65.21%) in group 2 having persistence of contrast in the PCS even after 20 min of Lasix injection, 3 (12.5%) units in Group 1 and 5 (21.79%) units in Group 2 having partial contrast retention at 20 min after lasix injection and 3 units in each group having nonvisualised kidneys.

Table 1

Comparison of mean differential renal function on ethylene dicysteine scan between the two groups before and after pyeloplasty (at 3 mo and more than 1 year after)

Group	Mean±SD (DRF range in %)			P
	Preoperative	Postoperative
		EC1	EC2	Pre-op	EC1	EC2
1	45.88±14.42 (5-80) (n=17)	42.64±9.62 (17-54) (n=16)	43.75±9.88 (17-58) (n=16)	0.117	0.584	0.385
2	39.22±9.75 (21-53) (n=18)	44.77±12.82 (20-68) (n=17)	42.25±8.56 (23-58) (n=16)
Total	42.45±12.52 (5-80)	43.74±11.26 (17-68)	43±9.13 (17-58)

DRF: Differential renal function; EC: Ethylene dicysteine; EC1: Renal scan at 3 mo postoperative; EC2: Renal scan at 1 year or more postoperative; SD: Standard deviation

There was no significant difference in DRF between the 2 groups postoperatively at 3 months or ≥1 year [Tables 1 and 2]. One patient each in Group 1 and 2 were lost to follow up after DJ stent removal. One patient in Group 2 with SUOD in EC1 did not undergo further EC scans in follow up. There was significant improvement in drainage between the pre- and post-operative EC scans (P < 0.001) in both the groups as per grading. There was no significant difference between the postoperative EC scan drainage pattern at 3 mo and ≥1 year in each group (Group 1, P = 0.93 and Group 2, P = 0.26). The number of patients with UOD and SUOD put together was marginally more in Group 1 at 3 months and ≥1 year after surgery compared to Group 2, which had a slightly higher number of patients with slow/delayed drainage at ≥1 year [Figure 2].

Table 2

Comparison of late ethylene dicysteine scan (>1 year follow-up) between the two groups for ±5% change in differential renal function

Change in differential renal function	Number of patients (%)		P
	Group 1 EC2 (n=16)	Group 2 EC2 (n=16)
>−5% reduction in function	1 (6.25)	3 (18.75)	0.60
±5% change in function	13 (81.25)	12 (75)	1
>5% increase in function	2 (12.25)	1 (6.25)	1

EC2: Renal scan at 1 year or more postoperative

Figure 2

Three dimension clustered column chart comparing % of patients in Groups 1 and 2 with unobstructed drainage, slow unobstructed drainage, and slow/delayed drainage at 3 months and more than 1 year after surgery on ethylene dicysteine scan

There was significant improvement within each group while comparing the preoperative and postoperative IVU function (Group 1, P = 0.007 and Group 2, P = 0.001). There was no significant difference between the two groups in the 3 months postoperative IVU function (P = 0.214).

Within each group, there was statistically significant improvement in the IVU drainage after surgery (P < 0.001). There was no significant difference (P = 0.990) between the two groups in postoperative IVU with both showing maximum IVU drainage by 10 min post Lasix, i.e., 14 (58.33%) units of Group 1 and 14 (60.86%) units of Group 2; two units from each group had drainage of contrast by 2 h or 5 min post Lasix injection; and 7 (29.16%) units from Group 1 and 4 (17.39%) from Group 2 had absence of contrast by 20 min post Lasix. There was 1 (4.16%) unit from Group 1 and 3 (13.04%) from Group 2 with faint persistence of contrast in PCS after 20 min of Lasix injection.

Pre- and postoperative APD of the renal pelvis on USG KUB were compared between the 2 groups in the late follow-up (mean 45.5 months). At a mean follow up of 47.2 ± 13.93 months (range 36–72) in Group 1 (n = 10), the pelvis size reduced from 25.33 ± 15.104 (7.3–53) mm to 8.8 ± 5.977 (5–23) mm (P = 0.013).At a mean follow up of 44.42 ± 18.46 months (range 18–67) in group 2 (n = 15), the pelvis size reduced from 24.6 ± 8.805 (15–40) mm to 13.866 ± 9.625 (5–40) mm (P = 0.011). The difference between follow up USG parameters of pelvis of the 2 groups was not significant (P = 0.077).

DISCUSSION

UPJO is the most common congenital abnormality of the upper urinary tract. Pyeloplasty has been originally described as an open procedure, but the minimally invasive approach is gaining acceptance.[781617181920] The point of interest lies in the difference between open procedures, wherein, the dismembered pyeloplasty usually includes a significant reduction of the extra renal pelvis, whereas in laparoscopic pyeloplasty often only the obstructed segment of the UPJ is resected prior to reconstruction, to reduce the operative time.[19] The functional outcome difference has been rarely studied.

In the follow up of UPJO patients, although extensively used, ultrasound alone may not be sufficient as residual hydronephrosis often persists without obstruction and interpretation is subjective. Dynamic renal scans and IVU can reliably detect the presence of obstruction and assess the outcome of surgical intervention.[1013]

Reismann et al. studied the relative kidney uptake and urinary drainage using123 J-orthoiodohippurate isotope renography in laparoscopic dismembered pyeloplasty without pelvic reduction and open dismembered pyeloplasty with extensive pelvic reduction and concluded that extensive resection of the renal pelvis is not necessary as there were no differences in the renographic outcome of comparable patients.[9] Our study also showed significant improvement in drainage in the postoperative EC scans, with function remaining stable in the majority (81.25% in Group 1 and 75% in Group 2) with no significant difference between the 2 groups.

Burgu et al. in their study of 42 UPJO patients reported that the resolution of APD in USG KUB was more prominent in the pelvic reduction group at earlier stages of the postoperative period, and that pelvic reduction significantly improved the renal washout time (T1/2) in MAG-3. DRF was not affected by pelvis reduction.[6] The preoperative APD of the patients in pelvic reduction group was 33.9 ± 5.1 mm and in the nonreduction group was 29.4 ± 3.7 mm. They concluded that excision of large portions of renal pelvis during pyeloplasty is not desirable and that when the renal pelvis is surgically not reduced, it reduces itself by 6 months. Morsi et al., in a study of 40 congenital unilateral UPJOs with an APD in the transverse plane of at least 4 cm, concluded that there was no significant difference between pelvic reduction and nonreduction pyeloplasty with regard to APD, split function and postoperative drainage.[12]

Our study has shown that drainage reaches normalcy faster when the pelvis is reduced although there was no effect on DRF and is similar to the study by Burgu et al. In long term follow up, the drainage patterns are similar although there is a slightly higher number of cases with slow drainage in Group 2. Although comparison between the 2 groups with respect to pelvic dimensions in late follow up on USG KUB may not be considered correct, we made 2 important observations. Almost 4 years after the surgery, the pelvis had reduced in size considerably in the nonreduction group also (P = 0.011). Although not statistically significant, (P = 0.077) the reduction with time was not equal to a surgically reduced pelvis and may be partly contributing to a slightly higher % of patients with slow drainage in long term follow up of Group 2.

Giant hydronephrosis, mostly secondary to UPJO, is a rare urological entity described as equivalent to one day's urine output, occupying a hemiabdomen, meeting or crossing the midline, and being at least 5 vertebrae in length. The significance of pelvis reduction would be more in giant hydronephrosis as the doubts of kinking and stasis would be more in them. Most of the literature on giant hydronephrosis is in the pediatric age group with patients usually undergoing nephrectomy or first stage pelvis drainage and second stage pyeloplasty.[20] In our study, we had seven patients with large pelvis (>5 cm) – 2 in Group 2 and 5 in Group 1. All showed improvement in drainage and function after pyeloplasty, regardless of the extent of pelvis reduction. There has been no study comparing reduction and nonreduction pyeloplasty in giant UPJOs. As our numbers are small, we feel that in patients with giant pelvis, a prospective study is required to bring to conclusion a comparison of these two techniques.

There were some limitations in the study. The number of patients was small (n = 42). These numbers reduced further for function comparison on EC scan as both groups had patients with single kidney and bilateral involvement. The age group range was wide, i.e., 2 months to 11 years although the mean age was similar in both the groups. Majority of the patients had preoperative APD pelvis of <5cm on USG KUB in both groups.

CONCLUSION

The outcome following dismembered pyeloplasty in moderate hydronephrotic UPJO with and without renal pelvis reduction was comparable with respect to function. Removing excess pelvis improves drainage significantly in the first few months after surgery whereas in the nonreduction group it appears to take a slightly longer period of time to reach normalcy. If the child is asymptomatic, and USG KUB is improving, postoperative renal scan for drainage may be done at 1 year compared to 3 months when pelvic reduction is not done so that parents are not unnecessarily alarmed.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.