A meta-analysis of the efficacy of ureteroscopic lithotripsy and extracorporeal shock wave lithotripsy on ureteral calculi

PURPOSE: To re-evaluated the clinic efficacy of ureteroscopic lithotripsy (URS) and extracorporeal shock wave lithotripsy (ESWL) on ureteral calculi with Cochrane systematic reviews in this paper. METHODS: We searched clinical randomized controlled trials and prospective controlled trials in databases such as Cochrane library, Medline, Springer, Elsevier Science Direct, PubMed. Pooled estimate of risk ratios (RRs), standard mean difference (SMD) with 95% confidence intervals (CIs) were used as measure of effect sizes. Summary effect estimates were also stratified by sample size, study design and study region. The overall effect sizes were derived using a random-effects model or fixed-effects model when appreciated, and meta-analysis were conducted with software RewMan 5.0. RESULTS: The meta-analysis suggested that there were significant differences of post-treatment stone free rate, repeat treatment rate, patients’ satisfaction, incidence of postoperative complications, operation time and hospital stays between ESWL treatment cases and URS treatment cases. But in the sample sizes analysis, there were no significant differences of the post-treatment stone free rate and repeat treatment rate when the sample sizes were less than 100. CONCLUSIONS: Compared to the ureteroscopic lithotripsy treatment, extracorporeal shock wave lithotripsy treatment provided a significantly lower post-treatment stone free rate, but it also obviously brought out less postoperative complications, shorter operation time and hospital stays.


Introduction
The presence of calculi in the urinary system is defined as urolithiasis, which represents the symptomatic manifestation of various metabolic disturbances, caused by a variety of pathological factors and their interaction 1 .Urolithiasis is a common and frequently-occurring disease with increased morbidity and high recurrence rate, including renal calculi, ureteral calculi, cystolith and urethral calculi, among which ureteral calculi is the most commonly one 2 .So far, ureteroscopic lithotripsy (URS) and extracorporeal shock wave lithotripsy (ESWL) are principle therapies in ureteral calculi treatment with satisfactory curative effect in the vast majority of cases, and both have its own advantages and disadvantages 3 .
ESWL, dates back to 1980s, is a clinically proved method of totally noninvasive, with low cost, few complications and no or shorter hospitalization [4][5][6] .URS is more invasive and done under general anaesthesia, but significantly better in terms of operative time, fluoroscopy time and time to achieve a stone-free state [6][7][8] .
URS is also better in broking hard stones and the ureter opening is simultaneously dilated by the scope to facilitate subsequent stone passage.In Middela et al. 9 study, ESWL was demonstrated an effective and minimally invasive method for treating ureteral stones.
While Hong and Park demonstrated that despite the liberal use of ESWL, ureteroscopic lithotripsy was still the preferred treatment modality for managing ureter stones at many hospitals and achieves an immediate stone-free state in a high percent of patients 10 .So it is still controversial which treatment is clinical preferred.
Recently, plenty researches compared URS and ESWL in ureteral calculi therapy, but obtained various results due to the differences in research design, recruit and exclusion criteria, measurement methods and so on.In this paper, we collected clinical prospective studies on URS and ESWL, and applied meta-analysis to evaluate the clinic efficacy of URS and ESWL comprehensively.

Source of material
We retrieved literatures in a systematic way from the Cochrane library, Medline, Springer, Elsevier Science Direct, PubMed and EMbase concerning URS and ESWL treating ureteral calculi with the retrieval deadline of December 2012.
The search terms included "ureteral stones", "ureteral calculi", "extracorporeal shock wave lithotripsy" and "ureteroscopy and randomized controlled trials".In addition, a manual search of citations from relevant original studies and review articles was performed for any additional studies.

Inclusion and exclusion criteria
Studies were included in the meta-analysis if they met the following criteria:

Data extraction and literature evaluation
The data were independently extracted and collected in unified forms from the primary studies by two reviewers.The data items included study details such as sample size, cases loss to follow-up and/or withdraw and research index/data.Disagreement was defused by discussion or the third researchers.
Jadad scale 11 was applied to assess the methodological quality of recruited clinical trials based on the extracted information such as study design, patients, and the clinical outcomes.The evaluations were also performed by two assessors independently.
The recruited studies were defined with high quality when their Jadad score were ≥3.

Statistical analysis
Meta-analysis was carried out by RevMan 5.0 provided by the Cochrane Collaboration.The point estimates of RR, SMD and their 95% CIs were calculated as effect sizes.We assessed the heterogeneity by testing Cochran's Q-statistic 12 and using I 2 = 100%×(Q−df)/Q 13 .A significant Q-statistic (P<0.05) or I 2 -statistic (I 2 > 50%) indicated heterogeneity were existed between studies, and then the random effect model should be used for meta-analysis.
Otherwise, heterogeneity is not significant and the fixed effect model Xu Y et al.
was applied.We further conducted subgroup analysis according to sample size (<100/≥100), study design (RCT / Non-randomized) and study region (Europe and America / Asia and Africa) to investigate the impacts of study characteristics on our outcomes.

Eligible studies and the characteristics
According to the above retrieval method, 829 potentially relevant trials in total were yielded.After screening their titles, abstracts and full publication reviewing, we removed reviews, case reports and retrospective studies.Finally, a total 13 English papers were eligible and recruited in our analysis [14][15][16][17][18][19][20][21][22][23][24][25][26] , whose general characteristics were shown in Table 1.The included studies were published between 1999 and 2013.A total of 2005 patients (998 ESWL treatment cases and 1007 URS treatment cases) were considered in the meta-analysis.The studies' sample size was between 13 and 210.In addition, there were 4 high quality studies with Jadad Score 3 and 9 studies with Jadad Score 2.

Meta-analysis of the stone free rate
All the 13 papers [14][15][16][17][18][19][20][21][22][23][24][25][26] reported the differences of stone free rate of patients after the treatment of ESWL and URS.The heterogeneity results (P<0.00001,I 2 =83%) indicated significant heterogeneities were existed between the studies, so we used the random effect model to evaluate the pooled estimates.The pooled RR (95% CI) was 0.82 (0.74-0.90) and the p-value of the test for overall effect was less than 0.0001 which demonstrated there was a significant 18% decrease of stone free rate of patients treated with ESWL compared to the URS surgeries (Figure 1).

Meta-analysis of the repeat treatment rate
A total of 10 papers [14][15][16][17][18][19][20][21]24,26 were analyzed with the repeat treatment rate of patients after the two different therapeutic methods. Therewas significant heterogeneities (P<0.00001,I 2 =85%) between the recruited studies and random effects model was applied for the analysis.The overall estimates of RR was 3.46 (95%CI: 1.50-7.97)and the test p-value for overall effect was 0.004 (Figure 2).It pointed out that there were significant differences of repeat treatment rate between the ESWL and URS.

Meta-analysis of the patients' satisfaction
There were only two papers 14,17 studied the patients' satisfaction after treatment with ESWL and URS.Fixed effects model were chose for the meta-analysis as the no-significant heterogeneity (P=0.85,I 2 =0.0%).In Figure 3, the total estimates (P=0.02,RR=0.87, 95%CI: 0.78-0.98)showed that patients' satisfaction in URS group was higher than ESWL group significantly.

Meta-analysis of the hospital stays
There were three papers 17,18,20 studied the hospital stays of patients with the treatment of ESWL and URS.The heterogeneity results (P=0.0001,I 2 =89%) indicated significant heterogeneities were existed between the studies, so random fixed effects model were applied to evaluate the pooled estimates.The pooled estimates (P=0.004,SMD=-1.71,95%CI: -2.88--0.55)demonstrated that the hospital stays in URS group was also significant longer than that in the ESWL group (Figure 6).

Meta-analysis of the operation time
There were three papers 17,18,20 analyzed the operation time of the two different therapeutic methods.Random effects model were chose due to the poor heterogeneity (P=0.04,I 2 =70%).The total estimates (P=0.002,SMD=-1.12,95%CI: -1.81--0.43)demonstrated that the operation time in URS group was significant longer than that in the ESWL group (Figure 5).

Subgroup analysis
The subgroup analysis of stone free rate and repeat treatment rate of ureteral calculi patients stratified by the sample size, study design and study region were showed in Tables 2 and   3, respectively.In the stratified analysis by the sample size, there were no significant difference of the stone free rate (RR=0.91;repeat treatment rate stratified by the study design and study region were consisted with the pooled estimates.

Discussion
It is well known that ESWL and URS are the main techniques in treating ureteral calculi.But the choice of ESWL or URS for ureteric stone management is one of the most commonly debated controversies in endourology, which may be partly attributable to parallel technology advancement in both fields 27 .In order to provide evidences for choosing which technique is better for patients, we applied meta-analysis to comprehensively evaluate the  25,28,29 .Our analysis was in accordance with these studies.
URS is also performed with longer operation time, hospital stays and more postoperative complications for its invasiveness.
Besides, Bierkens et al. 30 reported on complications following URS with small caliber instruments were decreased.And some studies also showed that the complication rate of URS was low and comparable with ESWL when the surgeries were in the hands of an experienced urologist 25,29,31 .Meanwhile, researches supported ESWL as the treatment of choice for ureteral stones as there were less complications and no need for major anesthesia with ESWL 32,33 .
The observed high heterogeneity in the pooled metaanalysis and the subgroup analysis indicated that the sample size, study design and study region affected the heterogeneity little.As the results of stone free rate and repeat treatment rate were not significant difference between the ESWL and URS when sample size <100, so the sample size could have an effect on the total estimates even it affected the heterogeneity little.
Nevertheless, there were certain limitations in our meta-analysis.Firstly, the sample size of some eligible trails is small; whether the eligible studies applied the blind method is not known; and the position and/or the size of the stone which might influence the treatment outcome are also unclear.differs, which might impact the outcomes.Therefore, further assessment with large sample sizes, clearly subgroup analyses (such as size and position of stones) and uniform follow-up information of postoperative complications should be carried out to evaluate the efficacy of ESWL and URS.

Conclusion
Extracorporeal shock wave lithotripsy performed with less postoperative complications, shorter operation time and hospital stays, and ureteroscopic lithotripsy provided better efficacy of higher post-treatment stone free rate, lower repeat treatment rate and higher patients' satisfaction on the ureteral calculi treatment.

( 1 )
research papers publicly published abroad; (2) research objects were adult patients with ureteral diseases definitely diagnosed by imaging; (3) the study design were clinical prospective study; (4) studies should include treatments of ureteroscopic lithotripsy (URS) and extracorporeal shock wave lithotripsy (ESWL); (5) data should be integrity, and the number of cases in different treatment groups as well as cases finished the trials should be explicit; (6) observation index should include authors, published journals and time, patients quantity and their operation time, hospital stays, rate of stone free post treatment, rate of repeat treatment and postoperative complications.Besides, references would be excluded when appeared the following conditions: 1) article was non-English literature; 2) paper was re-publication or the literature was used with same population data; 3) article was non-original literature such as review, letters and comments.

FIGURE 4 -
FIGURE 4 -Meta-analysis forest plots of postoperative complications in ureteroscopic lithotripsy (URS) treated group and extracorporeal shock wave lithotripsy (ESWL) treated group.

FIGURE 5 -
FIGURE 5 -Meta-analysis forest plots of operation time in ureteroscopic lithotripsy (URS) treated group and extracorporeal shock wave lithotripsy (ESWL) treated group.

FIGURE 6 -
FIGURE 6 -Meta-analysis forest plots of hospital stays in ureteroscopic lithotripsy (URS) treated group and extracorporeal shock wave lithotripsy (ESWL) treated group.

TABLE 1 -
General characteristics of the 13 eligible literatures/ trials involved.
NA: not available

TABLE 2 -
Subgroup analyses of stone free rate of ureteral calculi patients treated with ESWL and URS.

TABLE 3 -
Subgroup analyses of repeat treatment rate of ureteral calculi patients treated with ESWL and URS.