Dismembered Laparoscopic Anderson-Hynes Pyeloplasty Versus Nondismembered Laparoscopic Y-V Pyeloplasty in the Treatment of Patients with Primary Ureteropelvic Junction Obstruction: A Prospective Study

Introduction

E ven though some investigators recommend laparoscopic nondismembered procedures, laparoscopic dismembered pyeloplasty is considered the gold standard in the management of ureteropelvic junction obstruction (UPJO) by most urologists. ^{1 –6} As far as we know, only two reports have been published so far in which the results of laparoscopic dismembered Anderson-Hynes (A-H) and laparoscopic nondismembered Y-V pyeloplasties are retrospectively compared and statistically verified, ^4,7 Other studies do not provide the statistical verification of the analyzed data. ^6,8

The following report is a prospective study aimed to compare two laparoscopic techniques, A-H and Y-V, in terms of their therapeutic results and intraoperative and postoperative complications.

Patients and Methods

Fifty patients with primary UPJO who had undergone laparoscopic pyeloplasty between April 2008 and November 2010 were included in the study. The operations were performed consecutively by two experienced surgeons (TS, JK) Each of them had previously performed more than 40 laparoscopic pyeloplasties. The ethics committee of our university approved the study protocol. Having signed an informed consent, the patients were prospectively assigned randomly to laparoscopic A-H (even numbers) or Y-V (odd numbers) pyeloplasty. Each group included 25 patients.

In all patients, UPJO was diagnosed on ultrasonography, excretory intravenous urography (IVU), and diuretic renography (DR). Omnipaque (Amersham Health AS, Oslo, Norway) was used for IVU. In patients with a diagnosis of obstructed kidney on IVU, a four-grade scale was used to estimate the degree of hydronephrosis. ⁹ Grades 1 (mild) and 2 (moderate) were analyzed together. So were grades 3 (severe) and 4 (extreme).

For DR, Tc99m-EC (CSK, Lodz, Poland) was used. The data analysis included differential renal function, clearance half-time (T_1/2) after furosemide administration, and renogram curve patterns. The norm was defined as 45% to 55% contribution of each kidney to total renal function. A pyeloplasty was performed if the affected kidney contributed more then 15% to total renal function.

It was assumed that T_1/2 longer then 20 minutes was indicative of obstruction. In nonobstructed systems, T_1/2 was taken to be shorter than 12 minutes. Equivocal values were identified within the range from 12 to 20 minutes. The abnormal response to furosemide injection was recognized if it resulted in a plateau-like, gradually increasing or gradually decreasing renographic third phase.

The associated nonobstructing renal stones were recognized in five cases. In all patients, a visual analog scale (VAS) was referred to so as to determine the intensity of pain. The results of VAS are reported in Table 1. To assess overall renal function, serum creatinine concentration was investigated in every case. The preoperative data of both groups (Table 1) show no statistically significant differences, except for crossing vessels, more often observed in the Y-V group.

The operative technique was thoroughly described in our previous report. ⁷ Briefly, all patients underwent double pigtail stent (6F or 7F) placement before surgery. All procedures were performed using a transperitoneal approach through four ports. Crossing vessels, if identified, were isolated. Posterior crossing vessels were not observed in any case. Anterior crossing vessels were translocated cephalad and fixed to the edge of the Gerota fascia with three to four interrupted 3-0 polyglactin sutures (Johnson & Johnson Intl, St-Stevens-Woluwe, Belgium), irrespective of the type of plasty applied. In the course of A-H pyeloplasty, the stenotic segment was excised and the renal pelvis was trimmed if redundant (grade 3–4 hydronephrosis). The associated renal stones were removed from the renal pelvis with standard laparoscopic instruments. The ureter was spatulated, and ureteropelvic reanastomosis was performed. The preferred suturing material was 4-0 polyglactin with a curved needle.

In nondismembered Y-V pyeloplasty, two V-forming incisions were made on the anterior aspect of the renal pelvis. The incision was carried down the proximal ureter, traversing the area of stenosis. The anastomosis was performed, using interrupted sutures (4-0 polyglactin); the apex of the V-shaped flap was joined to the most inferior aspect of the ureterotomy, and then the anastomosis was completed over an internal stent. Above described, anterior Y-V pyeloplasty may not correct high ureteral insertion. Therefore classic Foley Y-V plasty was performed in such cases. ¹⁰

A 5-mm closed suction drain was left in place after completion of the anastomosis, and a 16F Foley catheter was left in the bladder. The drain was removed if the drainage was minimal (<50 mL/24h), and the Foley catheter was removed the next day. One month postoperatively, the stent was removed. Three months later, ultrasonography, DR, as well as the assessment of symptoms (VAS) were performed for all patients. Then, on each postoperative visit (every 3 months during the first 24 months after the operation and later once a year), the assesment of symptoms according to the VAS as well as ultrasonography was performed. DR was performed 13 and 25 months after the surgery.

Forty-four patients completed one diuretic renography at least. Six patients were lost to follow up. Thus they did not undergo the above examination. DR was equivocal (12 min <T_1/2 <20 min, stable differential renal function) for two patients who had presented with the flank pain before and were asymptomatic after pyeloplasty. IVU performed for them was equivocal, too (reduction of hydronephrosis but not visible ureteropelvic junction [UPJ]). Because a Whitaker test indicated normal renal pelvis pressure (lower than 15 cm H₂O), a decision was made to monitor the patients and provide DR every 6 months.

The longest follow-up period was 40 months for both the A-H group and the Y-V group; the shortest was 9 months for the A-H group and 10 months for the Y-V group. The mean follow-up was 26.2 months for the A-H group and 26.6 months for the Y-V group, the difference being not statistically significant (P=0.865).

Success was defined by three factors taken collectively: 80% or greater pain relief in comparison with the preoperative VAS score, no obstruction on DR (decreasing renographic excretion curve, T_1/2 <12 min), and improved or stable differential renal function.

The parameters studied in the two groups were the operative time, morbidity, hospitalization length, and objective outcome. The chi-square test and Fisher exact test were used to compare categoric values. Independent samples t test was used to compare numeric variables. The Shapiro-Wilks test was used to confirm the data to fit a normal distribution. A P value of less than 0.05 was considered statistically significant.

Results

All the procedures were completed with laparoscopy, with no need for open conversion. Crossing vessels (all anterior) were observed intraoperatively in 12 (48%) cases in the A-H group and in 21 (84%) in the Y-V group (P=0.017). The mean operative time (from cystoscopy to the last wound closure) for A-H was 172 minutes and for Y-V, 166 minutes (P=0.399). The mean hospitalization length was 4.4 days and 4.3 days for A-H plasty and Y-V plasty, respectively (P=0.342).

Intraoperative complications were observed in two patients. One was an injury of the crossing vessel, which was secured by Endo Clips. In the other patient, a double pigtail catheter was inserted intraoperatively because it had not been possible to insert a stent into the operated kidney before the surgery.

Postoperative complications, which developed in eight patients, were subdivided according to the Clavien classification of surgical complications. ¹¹ Prolonged urinary leakage (7 days after the procedure), which was observed in a patient in the Y-V group, was classified as a grade I complication. The only one grade II complication was fever of unknown origin observed in the patient with no urinary leakage, no abnormalities on physical examination, chest radiography and ultrasonography of the abdomen. The temperature dropped after a 3-day antibiotic treatment. There were five grade IIIa postoperative complications. In all these cases, fever, stent obstruction, and urinary leakage were observed. The obstructed stent was replaced; temperature went back to normal after treatment with antibiotics and antipyretics.

There was one grade IIIb complication. The patient presented to another clinic, 8 days after the operation previously carried out in our institution, with fever and flank pain. On admission, ultrasonography and CT indicated perinephric urinoma. Drainage of the urinoma was performed, and the temperature dropped within 2 days of antibiotic treatment. No significant difference was found in the incidence of complications between the two groups. Table 2 presents the operative data and complications.

The success rate was calculated in 44 cases. Six patients were lost to follow-up (three after A-H pyeloplasty and three after the Y-V procedure). Success was observed in 21/22 (95%) and in 19/22 (86%) patients after A-H and Y-V plasties, respectively. The difference between the success rates for the procedures was not found to be statistically significant (P=0.350) (Table 3).

Two patients for whom DR and IVU results were equivocal but the renal pelvis pressure was normal on a Whitaker test did not count as success. One of them completed a second diuretic renal scan, which revealed stable renal function and equivocal half-time to tracer clearance. The other patient has not undergone the second DR yet. In this case, however, ultrasonography has not indicated the worsening of hydronephrosis so far. Neither patient reported pain.

Both failures were observed in the Y-V group. The first patient did not have pain after the stent removal, but DR performed 3 months later showed a plateau-like third phase of the renogram curve with no furosemide effect. Endopyelotomy was recommended to the patient, but he refused to continue the treatment. The second patient was referred to our department 32 months after laparoscopic pyeloplasty because of left renal colic. DR revealed an abnormal renogram curve (gradually decreasing third phase with no furosemide effect), and half-time to tracer clearance was longer than 20 minutes. The patient underwent successfully antegrade endopyelotomy.

Because the difference in the number of cases with crossing vessels in the two groups was statistically significant, we analyzed the success rate for such patients. Even though the success rate in the A-H group was higher than in Y-V group (100% vs 83%), the difference was not statistically significant. Table 4 presents the above data.

Discussion

The anastomosis during laparoscopic A-H pyeloplasty is technically demanding because it necessitates considerable skill in intracorporeal suturing. In an attempt to make the operation easier and less time-consuming, some authors turned back to nondismembered procedures, reporting good therapeutic results. ^{4 –6} Our intention then was to undertake prospective research to compare two operative techniques: A-H plasty, commonly considered the gold standard, and nondismembered Y-V plasty. The procedures were performed by two urologists experienced in laparoscopy, which made it possible to minimize the influence of the learning curve of the participating surgeons on the outcome.

No statistically significant differences in operative times of the respective techniques were observed. We found it surprising because Y-V plasty seemed to us a subjectively easier procedure than A-H plasty. A retrospective data analysis of 90 patients who had undergone laparoscopic pyeloplasty at our institution from 2001 to 2007 indicated that the mean operative time for A-H plasty was considerably longer then for Y-V plasty. ⁷ Undoubtedly, crossing vessel translocation may affect the operative time. In our study, a crossing vessel was more often observed in the Y-V group than in the A-H group (a statistically significant difference), which might account for the comparable operative times of both procedures. The comparable operative times observed in the two groups may also result from the surgeons' proficiency in intracorporeal suturing—more sutures needed in A-H plasty did not necessarily prolong the operation.

Our findings show that the number of complications after both types of laparoscopic pyeloplasty were comparable. We were afraid that the injury of the crossing vessel could contribute to the failure of the procedure. After the operation, however, the patient was asymptomatic, and we did not observe obstruction on DR. The renal function was stable (no change in comparison with preoperative renal scan). Also, the blood pressure was normal during 24 months of follow-up. Similar outcomes of crossing vessel division were reported by Keeley and colleagues. ¹²

The most frequently occurring cause of postoperative complications in both groups was stent obstruction. To avoid stent obstruction, some researchers perform the operation without stent placement and report good therapeutic effects. ¹³ We have no experience in this respect, because only one patient in our early series (first 100 laparoscopic pyeloplasties performed from November 2001 to March 2008) had no internal drainage. Unfortunately, in this case, a percutaneous nephrostomy tube had to be placed because of a massive urinary leakage.

Most reports indicate that the mean hospitalization length after pyeloplasty varies from 1.2 to 7.2 days. ^6,14,15 Because a complications rate seems to be the main factor that may lengthen the hospital stay, we did not observe statistically significant differences in time of hospitalization in the compared groups. For A-H plasty and Y-V plasty, it was, respectively, 4.4 and 4.3 days.

One of the reasons that many urologists choose a dismembered procedure in the management of UPJO is the need of crossing vessel transposition. ^1,16,17 Rassweiler and associates ⁴ and Janetschek and coworkers ⁵ report good therapeutic results of nondismembered pyeloplasty in the case of an anterior crossing vessel, however. During the operation, the vessel is translocated cephalad and a Y-V plasty, which transposes caudally the newly created UPJ, increases its distance from the translocated vessel. On the basis of our own material involving open pyeloplasty, we observed that dorsal displacement of an anterior crossing vessel might not improve the anatomic situation and thus did not always give good therapeutic results. We found the same observation in the research of Janetschek and colleagues, ¹⁸ where it is claimed that dorsal displacement of anterior crossing vessels may even worsen the situation.

Cephalad translocation can be easily performed in dismembered and nondismembered plasties, and in our study, it was the only procedure in cases of anterior crossing vessels, irrespective of the type of procedure performed. It is noteworthy that only one patient from two failures in the Y-V group underwent cephalad crossing vessel translocation. In the second patient from our two failures, there was no crossing vessel; hence, we cannot assume that the failure of the Y-V procedure was caused by the method of crossing vessel displacement (cephalad translocation instead of dorsal transposition). Furthermore, in our patients with crossing vessels who were operated on using the A-H technique with cephalad translocation, the success rate was 100%. The above observations appear to indicate that cephalad translocation is an effective approach to crossing vessels.

We did not use any diagnostic techniques to assess the position of the translocated crossing vessels postoperatively. We relied only on the ultimate result of the operation. As mentioned above, in one of our two failures in the Y-V group, the crossing vessel was translocated during the operation. It cannot be excluded that the cause of the failure was the translocated crossing vessel, which fell to the original position postoperatively. Unfortunately, this patient refused to undergo the diagnostic procedures and to continue the treatment.

Few reports have been found comparing the results of laparoscopic dismembered and nondismembered pyeloplasties. ^4,6,7,8,15 Three studies report similar therapeutic results. ^4,6,7 Klingler and associates ⁸ reported a considerably lower success rate of Y-V plasty in comparison with A-H plasty, but only five patients constituted the Y-V group and all of them were operated on in the initial stages of the learning curve. Casale and coworkers ¹⁶ report the success rate of 94% for A-H plasty and 43% for Heineke-Mikulicz plasty, albeit in children.

Supporters of dismembered pyeloplasty indicate that although a nondismembered procedure can correct the ureteral stenosis, it cannot reduce the enlarged renal pelvis. ⁸ They also indicate that whereas nondismembered procedures can merely rearrange the dysplastic tissue in the UPJ, dismembered plasties are capable of excising it altogether. ¹⁴ Some authors, however, achieve good results without excising a stenotic segment during Y-V plasty. ^4,6 There is no evidence either that pelvic reduction improves the therapeutic results of laparoscopic plasty. ¹⁹

In the published series, the success rate of laparoscopic pyeloplasties ranged from 87% to 98%. ^4,16,20 The present study does not give the conclusive answer to the question whether laparoscopic A-H plasty and laparoscopic Y-V plasty are equally efficacious. Our analysis indicated a higher success rate for the A-H group (95%) than for the Y-V group (86%) The difference was not statistically significant. It is noteworthy, however, that both our failures were found in the Y-V group. The difference in the efficacy between the two procedures was even more conspicuous in patients with crossing vessels (100% vs 83.3%), but also in this subgroup, the difference was not statistically significant. Nevertheless, taking into consideration the statistical trend of our study (higher success rate in the A-H group), it seems that, until further results of other randomized studies comparing A-H and Y-V plasties appear, A-H plasty should remain the gold standard in the management of UPJO.

The number of patients in our series was relatively small, which could have affected the power of the study. It seems that it might be difficult to collect a large series of cases, however, using solid inclusion criteria, an accepted follow-up period, and a limited number of experienced surgeons. Multicenter studies could be helpful in increasing the sample size, but when evaluating the outcomes of such trials, differences in operative techniques and in the surgical experience of participating operators should be taken into consideration. We do believe that further randomized, single-center studies are needed to evaluate our preliminary findings.

Conclusions

In our prospective study, laparoscopic A-H plasty achieved a higher success rate then Y-V plasty, yet the difference was not statistically significant. The Y-V procedure seems to be subjectively easier than A-H plasty, but in experienced hands, the operative time and the number of intraoperative and postoperative complications are comparable.