Flexible Ureterorenoscopy for Renal and Proximal Ureteral Stone in Patients with Previous Ureteral Stenting: Impact on Stone-Free Rate and Morbidity

Introduction

Flexible ureterorenoscopy (f-URS) has become useful for the management of urinary calculi. Its efficacy and safety for all patients, regardless of their age, render f-URS an attractive option for treatment of stones less than 1.5 cm. ^1,2

However, a large number of patients required ureteral stenting (US) for renal colic or urinary sepsis before management of urolithiasis. This US has been proved to be efficient for relief of pain due to renal colic and for the management of sepsis in obstructed kidney. Ureteral stent may present side effects, such as lower back pain, dysuria, pollakiuria, or hematuria. ^{3 –7}

Many studies have investigated the utility of ureteral stent after f-URS, but few have analyzed its impact on stone-free rate (SFR) and morbidity of f-URS, except in children. ⁸ In this study, ureteral stent was placed in children to induce passive dilation of the ureteral meatus allowing effective ureteroscopy and no associated complications.

The primary goal of our study was to evaluate the impact of US, placed previously to the f-URS, on SFR according to stone location (renal or ureteral). The secondary objective was to evaluate the impact of US on technical aspects and complications of f-URS.

Patients and Methods

We conducted a retrospective study, including patients who underwent an f-URS for upper urinary tract (lumbar ureteral and renal) calculi between January 2004 and December 2010. Patients with pelvic or iliac ureteral lithiasis who underwent semirigid ureteroscopy were not included in this study. In total, 497 f-URS procedures in 373 patients were included and categorized into two groups, namely group 1, comprising 316 procedures in patients with a ureteral stent in place at the time of f-URS; and group 2, comprising 181 procedures in patients without ureteral stenting. As previously described, we considered each procedure as individual, and SFR analysis was performed on individual basis independent of the number of procedures performed in each patient. ^9,10

The following patient characteristics were recorded in patient files and central patient database in our Department: age, weight, height, body mass index (BMI), and sex ratio. Lithiasis characteristics (localization, size, and type), discovery circumstances (low back pain, hematuria, urinary infection, and obstructive pyelonephritis), procedural characteristics (ureteral dilatation, fragment extraction, and use of access sheath), and outcomes (length of stay, morbidity rate according to the Clavien-Dindo grading system, and 6 months-SFR) were compared between groups.

The indication for previous ureteral stent in group 1 was renal colic resistant to medical therapies, obstructive pyelonephritis, or other reasons (anuria and acute renal failure). Most of these patients were referred to our tertiary center for lithiasis management after imperative US by outside urologists, explaining why US duration, as well as stent size, was not standardized. Patients managed in our center underwent 7F silicon Double-J ureteral stenting (Coloplast, Fredensborg, Denmark).

All f-URS procedures were performed by experienced surgeons, using a standardized protocol, as previously described. ^9,10 Briefly, procedures required general anesthesia. Patients were installed in URS position. A rigid cystoscopy allowed exploration of the bladder and insertion of a stiff guidewire in the kidney. For patients with ureteral stent, the stiff guidewire was inserted and pushed to the kidney through the distal part of the stent pulled to the urethral meatus. Ureteral dilation was used when necessary, and standard access sheath was frequently used (especially in case of lower pole stone or high stone burden). Fluoroscopic guidance controlled the access sheath (12F–14F) insertion. The flexible ureterorenoscope (Flex-X²; Karl Storz, Tubingen, Germany) was then introduced into the ureter or kidney on the guidewire or through the access sheath. Renal cavities were inspected and when small stones (<5 mm) were identified, “en bloc” removal without laser fragmentation was performed. Larger calculi required fragmentation with a holmium:YAG laser (200, 272, or 365 μm fibers). Fiber size and laser energy settings were chosen by the surgeon based on stone size and location. After laser fragmentation, fragments were removed with a Nitinol stone basket. According to procedural conditions and surgeon's discretion, an ureteral stenting with a double pigtail indwelling ureteral stent (7F) or ureteral stent (8F) was used. Adjuvant medical expulsive therapy was not used in our center, but few patients may have received it based on surgeon decision.

Complications of f-URS (flank pain, hematuria, and sepsis) were described according to the Clavien-Dindo surgical complications grading system. US morbidity evaluation was not the objective of this study and was not analyzed.

SFR was defined by stringent criteria. We considered “stone free” as the complete absence of lithiasis fragments on CT-scan or on plain X-ray coupled with ultrasonography. Presence of large residual stone requiring additional treatment or nonexpulsed small residual fragments after 6 months was considered as a failure.

Results

Four hundred ninety-seven f-URS were done in our institution between 2004 and 2010. Among these procedures, 316 were performed in patients with ureteral stent placed 65.1 ± 53.4 days before f-URS (group 1) and 181 in patients without (group 2).

Age (51.5 ± 1 vs 50.6 ± 1 years, p = 0.51), gender (37.9% vs 43.3% female, p = 0.24), and BMI (25.8 ± 5 vs 25.7 ± 4.8 kg/m², p = 0.91) did not differ between the groups 1 and 2 (Table 1).

The indication for previous ureteral in group 1 was as follows: 51% (161) renal colic resistant to medical therapies, 26.5% (84) obstructive pyelonephritis, 5% (16) for other reasons (anuria and acute renal failure), and 12.5% (39) after a previous incomplete f-URS with residual stone requiring further f-URS. Sixteen patients (5%) underwent US for unknown reasons.

Both groups were not different for stone characteristics such as size, number, and type, but stone location was different between the two groups. Ureteral location was significantly more frequent in the group 1 (30.2% vs 16.3%, p = 0.0006) and renal location was higher in group 2 (87.1% vs 75.2%, p = 0.0016) (Table 1).

The mean operative time was, respectively, 95.2 ± 2.4 minutes in the group 1 and 99.7 ± 3.6 minutes in the group 2 (p = 0.24). Less ureteral dilatation was required in the group 1 than in the group 2 (8.8% vs 16.9% in the group 2, p = 0.001), but the use of ureteral access sheath was not different between the groups (77% vs 74.3%, p = 0.49). Monobloc extraction rate was similar in both groups (22.5% vs 16.6%, p = 0.11) (Table 2). Post f-URS stenting rate was similar in both groups (92.2% vs 91.1%, p = 0.49).

Imaging assessment of SFR was not homogenous during the recruitment period, but there was no statistical difference in CT-scan use between the both groups.

Global SFR (excluding two patients in group 1 and one in group 2 because of the absence of data about SFR) tended to be slightly higher in group 1 than in group 2 (72% vs 63%, p = 0.05). The SFR was impacted by stone size. For stone <10 mm, SFR tends to be nonsignificantly higher in group 1 (85.4% vs 76.9%, respectively, p = 0.083). For larger stone, SFR was not different between groups 1 and 2 (58.6% vs 46%, p = 0.112, for stone from 10 to 20 mm and 16.7% vs 38.5%, p = 0.192, for stone >20 mm). SFR for multiple location stone was 62.5% in group 1 and 48.9% in group 2 (p = 0.147).

Because of the discrepancy of ureteral location in both groups, we performed a subgroup analysis according to stone location (ureteral vs renal). SFR for lumbar ureteral location was significantly higher with a previously placed ureteral stent (81.5% vs 59.4%, p = 0.023). We conducted the same analysis for kidney location and did not find such results. Ureteral stent did not impact SFR whatever the renal stone size. For renal stone less than 10 mm, SFR was 83.6% in group 1 and 76.1% in group 2 (p = 0.188). For renal stone from 10 to 20 mm, SFR was, respectively, 56% and 46.9% in the two groups (p = 0.318) and 12.5% in group 1 and 45.5% in group 2 (p = 0.064) for stone higher than 20 mm. Finally, SFR for multiple kidney locations was not impacted by ureteral stent (57.8% in the group 1 vs 44.4% in group 2, p = 0.235).

No statistical difference was found for the surgical complications between the two groups. Global morbidity was comparable for the two groups (10.7% vs 11.8%, p = 0.70). The rate of Clavien-Dindo grade I (2.5% vs 4.5%, p = 0.23) and grade II (7.9% vs 6.9%, p = 0.66) complications was similar in both groups. No grade III complication was reported, but one complication grade IV was observed in group 1 (urinary sepsis requiring ICU).

In the multivariate analysis, using logistic regression, we analyzed factors associated with f-URS failure. Only stone size >10 mm (OR 4.69, 95% CI [3.04, 7.22], p < 0.0001) and multiple stone locations (OR 2.04, 95% CI [1.25, 3.33], p = 0.004) were significantly correlated with f-URS failure. Presence of ureteral stent at the time of f-URS did not impact SFR in this multivariate analysis (OR 0.72, 95% CI [0.46, 1.10], p = 0.13) (Table 3).

Discussion

In our study, ureteral stent was not associated with a better SFR, except in case of ureteral lithiasis, by univariate analysis. However, by multivariate analysis, only stone size and number were independently associated with f-URS failure. Complication rate of f-URS was not impacted by prior ureteral stent.

Ureteral stent before ureterorenoscopy was initially proposed in pediatric surgery. The objective was to perform passive dilation of the child's ureter to facilitate ureteroscopy. ⁸ Hubert and Palmer demonstrated the feasibility of this approach in 2005 in 26 young patients in whom the ureteral orifice was not amenable to endoscope progression. Ureteral stenting for 2 to 8 weeks allowed ureteroscopy to be effectively performed in all cases. In contrast, Tiryaki and colleagues ¹¹ evaluating the factors affecting outcomes of ureteroscopy in children did not show any effect of prior ureteral stenting on SFR.

In adult patient, few studies have investigated the utility of US patients before ureterorenoscopy. ^{12 –15} Most of these studies showed results similar to those presented in our work. Netsch and colleagues compared the impact of ureteral stent to a matched control population without stent and demonstrated a significantly higher SFR in patients with ureteral stenting (95.1% vs 86.7%, p = 0.013). ¹² Similar findings were also observed by Lumma and colleagues, who reported an increased rate of SFR after US (72.2% vs 59.4%). ¹³ This trend toward better results was also observed by Shields and colleagues, although without reaching statistical significance (88.7% vs 83.1%, p = 0.254). ¹⁵ By multivariate analysis, the same authors showed that stone size and number were the only factors associated with URS success. ¹⁵ These findings are confirmed by our results, as we also observed that ureteral stent was not an independent predictor of f-URS outcome, contrary to stone size and number, which were both shown to be significantly associated with f-URS failure.

One interesting finding of our study is the considerable impact of US on the treatment of lumbar ureteral stones. The SFR increased from 59.4% in the absence of ureteral stent to 81.5% in case of US (p = 0.02) in patients with ureteral stone location. These results are also in line with previous reports. ^{12 –14} Lumma and colleagues distinguished SFR for proximal (mid/proximal ureter and renal pelvis) and distal (distal ureter) locations. ¹³ In their study, stent did not impact URS results for the distal ureter, but positively impacted the SFR for proximal locations (67.1% with US vs 34.5% without US). These results are concordant with our results. In our study, only proximal (lumbar) locations were included and we demonstrated a positive impact of ureteral stent. In this regard, we should underline the existence of a possible recruitment bias in our study due to the high rate of ureteral location in our stented group. In our study, ureteral stent was performed in case of stone complications (renal colic, obstructive pyelonephritis, and acute renal failure), which are more frequent with ureteral stone locations. Nevertheless, the subgroup analysis of ureteral location showed a positive impact of ureteral stent on SFR after f-URS, although our study was not designed to investigate this particular group and, therefore, is not sufficiently powered to draw any firm conclusions.

The effect of US on the f-URS morbidity is another point of debate. We observed a similar rate of complications (around 10%) in patients with and without stent. Most of these complications were considered as minor, in accordance with the Clavien-Dindo classification. Only one major complication (grade IV) was observed in the entire cohort. Our results confirm those previously published showing that ureteral stent does not significantly impact on URS morbidity. ^{12 –14} Only Lumma and colleagues demonstrated that ureteral stent allowed a lower rate of complications for stones of proximal ureteral. ¹³

Contrary to our findings, major complications were reported after URS in few studies, especially perforation of the ureter. ^{12 –14} However, in all these studies, both semirigid ureteroscopes and flexible scopes were used. Our cohort is more homogenous and included only f-URS, possibly explaining the absence of ureteral perforation. Moreover, our protocol favored the wide use of access sheaths allowing protection of the kidney and the ureter and increase in SFR, ¹⁶ even if in nonstented patient, ureteral access sheaths may induce ureteral injuries. ¹⁷ Although our study was not designed to analyze complications due to ureteral access sheath, no serious ureteral injury was reported. The safety of ureteral sheaths' access was recently confirmed in a large retrospective cohort. ¹⁸ In our study, we reported that ureteral dilation to facilitate ureteral access sheath placement was more frequently required in nonpresented patients.

Chu and colleagues reported that pre-URS stenting was cost-effective for effective ureteroscopic treatment of stones measuring >1 cm. They hypothesize that this decrease of cost was due to a decreased operating room time and reoperative rate. ⁴ This decrease in operating time was not replicated in our study. ¹² Conversely, Lumma and colleagues found that procedural duration was increased by about 5 minutes in prestented patients. ¹³

Despite its retrospective design, our study presents several strong points. To our knowledge, this study presents the largest cohort to date to analyze the effect of ureteral stent on the efficacy and safety of f-URS. Moreover, contrary to many previous published reports, our series is homogeneous, including only f-URS, whereas other studies mixed both flexible and semirigid ureteroscopy. The number of surgeons involved in this study, as well as the absence of standardization (fiber size, laser energy settings), which could introduce bias in the results, could be criticized. However, in our view, this is more likely to reinforce the results of this “real life” cohort of patients treated in a University tertiary center. The SFR and morbidity of each procedure was performed on individual basis analysis. Some patients (n = 20) with two f-URS procedures may have one procedure included in each group depending on stent placement. This may also introduce a bias, as those failed cases may have more difficult stones (such as hard or in some unreachable area) and decrease SFR especially in group 1.

The discrepancy between both groups regarding stone location could also be a bias impacting results of SFR. To analyze this potential bias, we performed a subgroup analysis, including ureteral and renal location separately. We observed an increased SFR in patients with ureteral stent for ureteral location, but not for renal location.

The inhomogenicity of imaging follow-up could also be criticized. During the large analysis period of this retrospective study, we observed the increased use of CT-scan for management of urinary lithiasis. ¹⁹ We could hypothesize that the higher sensitivity of CT-scan compared with plain X-ray and ultrasonography allowed detection of more residual fragments leading to decrease of the SFR. However, the use of CT-scan was well balanced between the groups minoring its impact.

One of the main limitations of our study is that we were unable to analyze the impact of procedural duration or the different stent models and diameters. The impact of such factors on passive ureter dilation and f-URS success remains unknown. However, the retrospective design of the study, as well as the high number of patients undergoing ureteral stent by outside urologists before subsequently being referred to our university center for f-URS, precluded any such analysis.

Conclusion

Our study shows that technical aspects, such as duration and complications of the f-URS procedure, were not modified by ureteral stenting, except the need for ureteral dilatation. f-URS for stone retrieval in patients with US was not associated with a better SFR. By multivariate analysis, ureteral stent was not found to be independently associated with f-URS outcome. The morbidity of f-URS was not affected by the presence of US before f-URS.