The Accordion Antiretropulsive Device Improves Stone-Free Rates During Ureteroscopic Laser Lithotripsy

Abstract

Purpose:

The Accordion is a novel endoscopic device that prevents retropulsion of ureteral stones and their fragments during ureteroscopic laser lithotripsy. We describe our experience with its use focusing on three main endpoints: operating time, fluoroscopy time, and stone-free rates.

Methods:

Of 308 consecutive cases of unilateral ureteroscopic laser lithotripsy from 2006–2010, we analyzed 235 cases of ureteral stones. Chart review was performed to document patient demographics (age, sex, and race), stone characteristics (stone size, density, laterality, location, and multiplicity), operative characteristics (use of preoperative and/or postoperative stents, ureteral balloon dilators, ureteral access sheaths, the Holmium laser, and the Accordion device), and surgical outcomes (operative time, fluoroscopy time, stone-free status, and complications).

Results:

The baseline characteristics between the Accordion and non-Accordion group were statistically similar. In univariate nonparametric tests of medians, Accordion device usage was not associated with a significant reduction in fluoroscopy time (median 1.68 vs. 1.95 minutes, p=0.28) or operating time (median 52.5 vs. 61 minutes, p=0.19). However, the stone-free rate for the Accordion group was significantly higher compared to the non-Accordion group (84.2% vs. 53.6%, p=0.001). In multivariate generalized linear models, Accordion usage was not associated with decreased operating or fluoroscopy times. Accordion use was associated with statistically significant greater odds of stone-free status (odds ratio 4.35, 95% confidence interval 2.36–8.00). Complication severity and rates were comparable between the two groups.

Conclusions:

The Accordion antiretropulsive device improves stone-free rates during ureteroscopic laser lithotripsy. Prospective studies are needed to validate these results.

Introduction

U reteroscopic lithotripsy is a useful tool in the urologist's armamentarium for treatment of ureteral and renal stones. Stone retropulsion during ureteroscopic lithotripsy of ureteral stones can be frustrating and potentially lead to increased radiation exposure from additional fluoroscopy, longer operating times, lower stone-free rates, need for additional procedures, and ultimately, higher healthcare costs. It has been reported that up to 15% of distal ureteral stones and 48% of proximal ureteral stones migrate proximally during ureteroscopy.^1

–4 Several commercially available devices have been developed to prevent proximal migration of ureteral stones during ureteroscopic lithotripsy. These include the Accordion® (PercSys, Palo Alto, CA), the Stone Cone^® (Boston Scientific, Natick, MA), and the NTrap^® (Cook Medical, Bloomington, IN).^{3

–11} In general, each of these devices acts as a backstop once deployed within the ureter and is connected to a wire or catheter. Newer technologies include reverse thermosensitive polymers, such as the BackStop^® (PluroMed, Woburn, MA), a liquid at temperatures below 16°C, and a viscous gel at body temperature. It is applied proximal to the ureteral stone before starting lithotripsy, and then dissolved with cool irrigation at the termination of the case.¹² Proponents of this technology state that it avoids the need for an additional wire or catheter to remain within the ureter during ureteroscopy thereby allowing for greater maneuverability of the ureteroscope and reducing the possibility of ureteral injury. However, relatively little published literature explores the utility of such devices and their effect on operating times and other parameters.

Materials and Methods

After approval from our institutional review board, we identified 308 consecutive cases of unilateral ureteroscopic laser lithotripsy between October 2006 and May 2010. We limited the analysis to cases performed for ureteral stones, since antiretropulsive devices are not indicated for stones in the kidney. These cases were performed by three surgeons. Holmium laser lithotripsy was used in all cases unless the stones were small enough to be removed with an endoscopic basket, in which case, no laser lithotripsy was performed. Retrospective chart review was performed to document patient demographics, including age, sex, race, stone characteristics (stone size, density, laterality, location, and multiplicity), operative characteristics (use of preoperative and/or postoperative stents, ureteral balloon dilators, ureteral access sheaths, the Holmium laser, and the Accordion device), and surgical outcomes (operative time, fluoroscopy time, stone-free status, and complications). Stone-free status was ascertained by postoperative X-ray or computed tomography scan performed within 3 months of the procedure, depending on the radiographic characteristics of the stone. Accordion use was determined by surgeon preference.

The end points of operating and fluoroscopy time were defined as continuous variables. Univariate comparisons between the Accordion and non-Accordion groups were performed using the nonparametric Wilcoxon rank-sum test. The end point of stone-free status was defined as a binary variable and the corresponding comparisons were performed using the chi-square or Fisher's exact test.

Multivariate analysis was then used to evaluate the effect of Accordion use on operating time, fluoroscopy time, and stone-free status after ureteroscopy. In these analyses, Accordion use was defined as a binary variable (yes/no). Other predictors included patient age (years), race (white, hispanic, Asian, black, other), stone size (diameter in the largest dimension in mm), stone location within the ureter (proximal, mid, distal), the presence of a previously placed ureteral stent (yes/no), and use of a ureteral balloon dilator (yes/no), ureteral access sheath (yes/no), Holmium laser (yes/no), and postoperative stent (yes/no). Using generalized linear models, unadjusted effect estimates of Accordion use were first obtained in univariate regression. Then, a second model was generated adjusting for patient age, race, and gender. A final model containing all the preoperative characteristics above was then generated.

Statistical analysis was performed using SAS 9.2 (SAS Institute, Cary, NC). Tests for significance were two-sided and a p-value≤0.05 was considered statistically significant.

Results

Study cohort

Of the 308 patients, there were 235 (76.3%) cases performed for ureteral stones. Of these cases, the Accordion device was used in 76 (32.3%) cases. The characteristics of our cohort are listed in Table 1. There were no statistically significant differences between the baseline demographics, including age, gender, and race between the Accordion and non-Accordion group. The Accordion device was more likely to be used for distal ureteral stones (p=0.03). The ureteral access sheath was less likely to be used in the Accordion group (p=0.0007) and the Holmium laser was more likely to be used in the Accordion group (p=0.02).

Table 1.

Patient Characteristics

	Accordion	Non-Accordion	p-Value
Patients, n (%)	76 (32.3%)	159 (67.7%)
Median age, year (IQR)	44.6 (33.6–52.9)	46.5 (33.9–58.6)	0.34
Gender, n (%)
Male	29 (38.1%)	82 (51.6%)	0.06
Female	47 (61.8%)	77 (48.4%)
Race, n (%)
Hispanic	35 (46.1%)	72 (45.3%)	0.89
White	23 (30.3%)	46 (28.9%)
Asian	14 (18.4%)	27 (17.0%)
Black	2 (2.6%)	5 (3.1%)
Other	2 (2.6%)	9 (5.7%)
Multiple stones, n (%)
Yes	16 (21.1%)	29 (18.2%)	0.61
No	60 (78.9%)	130 (81.8%)
Median stone size, mm (IQR)	7 (6–10)	7 (5–9)	0.06
Stone side, n (%)
Left	34 (44.7%)	79 (49.7%)	0.48
Right	42 (55.3%)	80 (50.3%)
Stone location in ureter, n (%)
Proximal	25 (32.9%)	78 (49.1%)	0.03
Mid	20 (26.3%)	24 (15.1%)
Distal	31 (40.8%)	57 (35.9%)
Median stone density, HU (IQR)	856 (669–1026)	754 (512–1030)	0.08
Preoperative stent, n (%)
Yes	22 (29.0%)	34 (21.4%)	0.20
No	54 (71.0%)	125 (78.6%)
Preoperative hydronephrosis, n (%)
Yes	61 (80.3%)	133 (83.7%)	0.52
No	15 (19.7%)	26 (16.4%)
Ureteral balloon dilation, n (%)
Yes	6 (7.9%)	20 (12.6%)	0.28
No	70 (92.1%)	139 (87.4%)
Ureteral access sheath, n (%)
Yes	21 (27.6%)	81 (50.9%)	0.0007
No	55 (72.4%)	78 (49.1%)
Holmium laser, n (%)
Yes	73 (96.1%)	138 (86.2%)	0.02
No	3 (3.9%)	22 (13.8%)
Postoperative stent, n (%)
Yes	70 (92.1%)	139 (87.4%)	0.28
No	6 (7.9%)	20 (12.6%)
Median operating time, minutes (IQR)	52.5 (36–83.5)	61 (41–91)	0.19
Median Fluoroscopy time, minutes (IQR)	1.68 (1.28–2.51)	1.95 (1.32–2.80)	0.28
Stone-free, n (%)
Yes	32 (84.2%)	45 (53.6%)	0.001
No	6 (15.8%)	39 (46.4%)
Complications, n (%)
Yes	1 (1.3%)	2 (1.3%)	1.00
No	75 (98.7%)	157 (98.7%)

IQR=interquartile range.

Primary endpoints

In univariate nonparametric tests of medians, the use of the Accordion device was not associated with a significant reduction in fluoroscopy time (median 1.68 vs. 1.95 minutes, p=0.28) or operating time (median 52.5 vs. 61 minutes, p=0.19). However, the stone-free rate for the Accordion group was significantly higher for the Accordion group (84.2% vs. 53.6%, p=0.001). In multivariate analysis using generalized linear models to identify statistically significant predictors of fluoroscopy time and operating time, Accordion use was not associated with decreased times even when adjusted for patient demographics, stone characteristics, or operative characteristics. In multivariate logistic regression, Accordion use was associated with statistically significant greater odds of stone-free status after ureteroscopic lithotripsy (odds ratio 4.35, 95% confidence interval 2.36–8.00). These data are summarized in Table 2 –4.

Table 2.

Effect of Accordion Use on Operating Time ^a

Model	Beta	SE	p-Value
Unadjusted	0.092	0.075	0.22
Adjusted for age, race, gender	0.065	0.075	0.38
Adjusted for preoperative characteristics^b	0.057	0.066	0.39

Operating time was log transformed to meet normality assumptions of generalized linear models.

Multivariate adjustments for age, race, gender, stone size, stone density, stone side, stone location in the ureter, preplaced stent placement, postoperative stent placement, use of ureteral balloon dilation, use of ureteral access sheath, and use of Holmium laser.

Table 3.

Effect of Accordion Use on Fluoroscopy Time ^a

Model	Beta	SE	p-value
Unadjusted	0.079	0.084	0.35
Adjusted for age, race, gender	0.051	0.084	0.54
Adjusted for preoperative characteristics^b	−0.001	0.082	0.98

Fluoroscopy time was log transformed to meet normality assumptions of generalized linear models.

Table 4.

Effect of Accordion Use on Stone-Free Rate ^a

Model	Beta	SE	p-Value
Unadjusted	1.53	0.49	0.002
Adjusted for age, race, gender	1.57	0.51	0.002
Adjusted for preoperative characteristics^b	1.47^c	0.61	0.01

Risk of residual stone was modeled as a binary variable using logistic regression.

This corresponds to an odds ratio of 4.35 (95% confidence interval, 2.36–8.00) for stone-free status in the Accordion group.

Secondary endpoints

All the complications in this series were minor ureteral injuries and none required an unplanned hospitalization or readmission. There was one complication in the Accordion group and two complications in the non-Accordion group. Because of the small numbers, the Fisher's exact test was used to compare the complication rates between the two groups, and the two-sided p-value was 1.00.

Discussion

This is the first study to our knowledge analyzing the clinical safety and efficacy of the Accordion antiretropulsive device. Although using the Accordion device did not significantly reduce fluoroscopy or operative times, it did improve the stone-free rate by 30.6% in univariate analysis. In multivariate logistic regression, the odds ratio of being stone-free when the Accordion device was used was 4.35.

One of the difficulties of retrospective, nonrandom studies is selection bias. However, upon examination of the demographics and stone characteristics of our cohort, the Accordion and non-Accordion groups appear to be reasonably matched. There were no significant differences between the age, gender, and race of the two groups. In terms of stone characteristics, the only statistically significant difference was a higher proportion of patients in the Accordion group with distal ureteral stones. Our surgeons regularly use flexible ureteroscopy via ureteral access sheaths to manage proximal stones and chasing migrated stone fragments would not necessarily add too much time to the procedure with this type of approach. On the other hand, we preferentially manage distal stones with semirigid ureteroscopy during which retropulsion of stones and stone fragments can be much more problematic as that can require the conversion to flexible ureteroscopy to access the more proximal ureter and renal pelvis and lead to instrumentation of the ureter multiple times.

We also noted that our surgeons were less likely to use the ureteral access sheath and more likely to use the Holmium laser for patients in the Accordion group. The former finding reflects our preference for managing more distal ureteral stones with semirigid ureteroscopy as we find that the semirigid ureteroscopy equipment provides a better optical image compared to flexible ureteroscopy equipment at our institution. Our group uses only the Holmium laser for ureteroscopic lithotripsy. In cases where the Holmium laser was not used, the stones were small enough to be removed by an endoscopic basket placed ureteroscopically without the need for lithotripsy. One possible explanation for the higher rate of Holmium laser use in the Accordion group is that our surgeons may feel more comfortable performing lithotripsy when they do not have to worry as much about stone retropulsion. In cases in which the Accordion device was not available for use, we may have preferentially tried to basket out the stones rather than risk performing lithotripsy and causing proximal migration of the stone and its fragments. Stone size was not a likely factor contributing to this difference, as this variable was similar in both groups.

Although we did find a significant association between Accordion usage and stone-free rates on postoperative imaging, our stone-free rates may be biased since we did not routinely perform postoperative imaging in the more straightforward cases where there was only one stone on preoperative imaging and there were no difficulties during the case. Nevertheless, a significant proportion of patients from each group underwent postoperative imaging (50.0% of the Accordion group and 45.2% of the non-Accordion group). Another limitation of our data is that we are unable to determine the need for additional stone procedures in each group. A conservative estimate can be attained by assuming that all those who were shown not to be stone-free postoperatively required an additional procedure (6 in the Accordion group versus 39 in the non-Accordion group). Other characteristics, such as Holmium laser settings and stone migration rates were not noted, so it is unclear if these factors contribute to stone-free status.

We noted two complications in the non-Accordion group and one in the Accordion group. This difference was not statistically significant. All three were small ureteral perforations that were treated with postoperative ureteral stenting and did not require an unplanned hospitalization or further invasive procedures. These complications arose from placing a guidewire past an impacted ureteral stone and not from placement of the Accordion device.

The Accordion device is a hydrophilic microcatheter-based tool with a film occlusion mechanism that acts as a ureteral backstop once deployed. A newer version has recently been developed, which allows coaxial deployment over a guidewire and may thereby obviate the need for a ureteral access sheath. Although there is an additional cost associated with using the Accordion device during ureteroscopic lithotripsy, a 30.6% improvement in stone-free rates may potentially offset this cost by reducing the need for further diagnostic tests and procedures.

It would be interesting to see if the benefits of such ureteral occlusion devices are replicated in a larger prospective randomized trial and if these benefits really translate into overall cost savings due to shorter operating times and a decreased need for additional stone procedures. Rane et al have recently studied the efficacy of the BackStop reverse thermosensitive polymer in preventing stone retropulsion (8.8% vs. 52.9%), but use of the polymer did not lead to improved stone-free rates or a reduced need for additional procedures.¹² However, with 34 patients each in the treatment and control groups, the study may have been underpowered to detect such a difference. Wang et al published results from a randomized controlled trial in which the NTrap device was used to prevent stone retropulsion of proximal ureteral stones larger than 9 mm.¹³ Although the NTrap group had a lower rate of stone retropulsion (0% vs. 12%, p=0.007), cases in which the device was used had a significantly longer operative time (40.50 minutes vs. 33.84 minutes, p<0.001). Furthermore, use of the NTrap device did not significantly reduce stone-free rates at follow-up (100% vs. 91.2%, p=0.098). With fewer than 60 patients in each group, this study was designed to have 80% power to detect a 30% difference between the two groups, so a smaller, but real difference may not have been detectable due to these limited numbers. Future studies comparing ureteral occlusion devices head-to-head in terms of measures of effectiveness and complication rates are still needed to identify if any one device is clearly superior to the others currently being used.

Conclusion

The Accordion device is an effective and safe endoscopic device that can improve stone-free rates. The added cost of this device may potentially be offset by the reduced need for additional procedures.

Footnotes

Acknowledgment

This study was approved by Santa Clara Valley Medical Center's institutional review board.

Disclosure Statement

R.S. is a member of the scientific advisory board of PercSys. All other authors do not have any competing financial interests to disclose.

Abbreviations Used

References

Chow

, Patterson

, Blute

et al. Ureteroscopy: Effect of technology and technique on clinical practice. J Urol, 2003; 170:111.

Dretler

. Preventing stone migration during ureteroscopy: The Stone Cone “rules.” Contemp Urol, 2006; 18:57.

Holley

, Sharma

, Perry

et al. Assessment of novel ureteral occlusion device and comparison with stone cone in prevention of stone fragment migration during lithotripsy. J Endourol, 2005; 19:200–203.

Eisner

, Dretler

. Use of the Stone Cone for prevention of calculus retropulsion during Holmium:YAG laser lithotripsy: Case series and review of the literature. Urol Int, 2009; 82:356–360.

Olbert

, Keil

, Weber

et al. Efficacy and safety of the Accordion stone-trapping device: In vitro results from an artificial ureterolithotripsy model. Urol Res, 2010; 28:41–46.

Pardalidis

, Kosmaoglou

, Kapotis

. Endoscopy vs extracorporeal shockwave lithotripsy in the treatment of distal ureteral stones: Ten years' experience. J Endourol, 1999; 13:161.

Ahmed

, Pedro

, Kieley

et al. Systematic evaluation of ureteral occlusion devices: Insertion, deployment, stone migration, and extraction. Urology, 2009; 73:976–980.

Chew

, Gotto

, Teichman

et al. The Accordion: A new device to prevent stone migration during ureteroscopic lithotripsy. J Endourol, 2007; 21,Suppl 1:A108.

Dretler

. The Stone Cone: A new generation of basketry. J Urol, 2001; 165:1593–1596.

10.

Wosnitzer

, Xavier

, Gupta

. Novel use of a ureteroscopic stone entrapment device to prevent antegrade stone migration during percutaneous nephrolithotomy. J Endourol, 2009; 23:201–207.

11.

Vejdani

, Eisner

, Pengune

et al. Effect of laser insult on devices used to prevent stone retropulsion during ureteroscopic lithotripsy. J Endourol, 2009; 23:705–707.

12.

Rane

, Bradoo

, Rao

et al. The use of a novel reverse thermosensitive polymer to prevent ureteral stone retropulsion during intracorporeal lithotripsy: A randomized, controlled trial. J Urol, 2010; 183:1417–1423.

13.

Wang

, Huang

, Chang

. Randomized trial of NTrap for proximal ureteral stones. Urology, 2011; 77:553–557.