Percutaneous nephrolithotomy versus open surgery for surgical treatment of patients with staghorn stones: A systematic review and meta-analysis

Objectives To compare the efficacy and safety of percutaneous nephrolithotomy (PCNL) and open surgery (OS) for surgical treatment of patients with staghorn stones based on published literatures. Materials and methods A comprehensive literature search of Pubmed, Embase, CNKI and Cochrane Library was conducted to identify studies comparing outcomes of PCNL and OS for treating patients with staghorn stones up to Jan 2018. Results There was no significant difference in final-SFR between PCNL and OS (odds ratio[OR]: 1.17; 95% confidence interval [CI]: 0.64, 2.15; p = 0.61), while PCNL provided a significantly lower immediate-SFR compared with OS (OR: 0.29; 95% CI: 0.16, 0.51; P < 0.0001). PCNL provided significantly lower overall complication rate, shorter operative times, hospitalization times, less blood loss and blood transfusion compared with OS (OR: 0.59; 95% CI: 0.41, 0.84; P = 0.004), (weighted mean difference [WMD]: -59.01mins; 95% CI: -81.09, -36.93; p < 0.00001), (WMD: -5.77days; 95% CI: -7.80, -3.74; p < 0.00001), (WMD: -138.29ml; 95% CI: -244.98, -31.6; p = 0.01) and (OR: 0.44; 95% CI: 0.29, 0.68; P = 0.00002), respectively. No significant differences were found in minor complications (Clavien I-II) (OR: 0.72; 95% CI: 0.47, 1.09; p = 0.12) and major complications (Clavien III-V) (OR: 0.5; 95% CI: 0.23, 1.08; P = 0.08). In subgroup analysis, there were no significant differences for overall complications and operative times between mini-PCNL and OS. In sensitivity analysis, there was no significant difference for overall complications between PCNL and OS. Conclusion Our analysis suggested that standard PCNL turns out to be a safe and feasible alternative for patients with staghorn stones compared to OS or mini-PCNL. Because of the inherent limitations of the included studies, further large sample, prospective, multi-centric and randomized control trials should be undertaken to confirm our findings.


Introduction
Staghorn stones still represent an intractable challenge to urologists. Open surgery (OS) was once considered to be the "gold standard" for the surgical treatment of staghorn calculi. However, thanks to miniaturization of endoscopic devices, increasing quality of optic systems, advent of holmium laser use and increasing experience in endoscopic surgery, the surgical management of staghorn calculi has been revolutionized. And the American Urological Association(AUA) guidelines for the management of staghorn calculi recommend percutaneous nephrolithotomy (PCNL) as the modality of choice and standard of practice [1]. However, not only the stone burden but also the morphology of stones can significantly affect the outcomes of PCNL in the management of staghorn calculi [2,3]. This is in contrast to OS, which is little affected by the morphometric index of staghorn calculi. Due to some reasons such as the unavailability of surgical instruments, higher stone free rate(SFR) or shorter operative times, many urologists still recommend OS for patients with complex staghorn calculi [4][5][6].
Some previous studies [6][7][8][9][10][11][12][13][14][15] had compared PCNL versus OS. Nevertheless, all these studies were small samples, and the results were controversial and inconclusive. The optimal treatment for staghorn calculi is still under debate. No meta-analysis has investigated the efficacy and safety of those two procedures. And whether PCNL is safer or more effective when compared to the OS remains unsettled. Therefore, to provide comprehensive information about the strategy of PCNL as well as OS in the treatment of staghorn renal stones, we performed this systematic review and meta-analysis of published studies comparing the efficacy and safety of PCNL and OS for surgical treatment of patients with staghorn renal stones. We hope it may guide urologists and patients to decide on the treatment modality, and to select the optimal treatment.

Materials and methods
A prospective protocol of objectives, literature-search strategy, inclusion and exclusion criteria, study selection, data extraction, outcome measurements, and methods of statistical analysis was prepared a priori according to the Preferred Reporting Items for Systematic Reviews and Meta-analysis [16].

Search strategy
A comprehensive literature strategy search was performed by two members (Chen and Zhang) independently in Jan 2018. The PubMed, Embase, CNKI and the Cochrane Library databases were used to identify relevant studies up to Jan 2018. Separate searches were done with the following search terms: "'percutaneous nephrolithotomy' or 'PCNL' or 'PCN'" and "'Open surgery' or 'OS'" in combine with "'staghorn calculi' or 'staghorn renal stones'".

Inclusion and exclusion criteria
The selected studies were included based on the following criteria: (1) studies reported comparison between PCNL and OS in patients with staghorn calculi; (2) the outcome measures consisted of at least two of the following things: complications, SFR, hospitalization times, operative times, blood loss, and blood transfusion. Exclusion criteria are as follows: (1) repeated publications or conference proceedings; (2) non-published materials, editorials or reviews; (3) studies containing patients with serious urinary infection, renal insufficiency, musculoskeletal deformities, solitary kidney or congenital abnormalities.

Study selection and data extraction
We screened the studies according to inclusion and exclusion criteria. Two authors (Feng and Yue) independently extracted data and appraised both quality and content. We contacted the authors of relevant studies to supplement incomplete data. Where disagreement arose, papers were re-examined and discussed, and the consensus was reached by the adjudicating senior authors (Zeng and Chen). The extracted data including: first author, year of publication, baseline patient characteristics, study period, study design, interventions, outcome measures, variations in PCNL techniques, statistical methods, and study conclusions. The outcomes included complications, SFR, hospitalization times, operative times, blood drop, and blood transfusion.

Quality assessment and statistical analysis
The level of evidence (LE) of all included studies was assessed by the criteria provided by the Oxford Centre for Evidence-based Medicine [17]. And the quality of non-randomized controlled trials (non-RCTs) was assessed by Newcastle-Ottawa Scale. [18] The Cochrane risk of bias tool was applied to assess the methodological quality of RCTs. [19] All the meta-analyses were performed using Review Manager 5.2 software. The odds ratio (OR) and weighted mean difference (WMD) were used to compare dichotomous and continuous variables, respectively. All results were reported with 95% confidence intervals (CIs). Chi-square test and I-square test were used for testing heterogeneity between studies. If heterogeneity was not significant (P > 0.10, I 2 < 50%), fixed-effect model was employed, otherwise, random-effect will be adopted. The results of the meta-analysis were expressed using forest graphs or tables. The Ztest determined the pooled effects, and P < 0.05 was considered statistically significant. Subgroup analyses were performed to compare standard PCNL and mini-PCNL with OS. Funnel plots were used to screen for potential publication bias. Sensitivity analysis was undertaken using studies of high quality. Only outcomes with three or more than three studies were included in the sensitivity analysis.

Sensitivity analysis
When high quality studies were assessed, no change in terms of the significance of each of the outcomes except for overall complications. Meta-analysis of 7 high quality studies [6][7][8][9][10]14,15] revealed that there was no significant difference between the two groups for surgical treatment of patients with staghorn stones, with no between-study heterogeneity. Between-study heterogeneity was significantly reduced by the sensitivity analysis for final SFR and blood loss. While heterogeneity remained statistically significant in operative times and hospitalization times. (Table 3)

Publication bias outcomes
Funnel plots were conducted to assess the publication bias in this meta-analysis that reported overall complications (Fig 6). All studies lie inside the 95% CIs, with an even distribution around the vertical, indicating no obvious publication bias.

Discussion
This meta-analysis of 10 studies, which included 921 patients from 1 RCT, 2 prospective casecontrol studies and 7 retrospective case-control studies, comparing the efficacy and safety of PCNL and OS for patients with staghorn stones. The pooled data showed that there was no significant difference in final-SFR between PCNL and OS, while PCNL provided a significantly lower immediate-SFR compared with OS. PCNL provided significantly lower overall complication rate compared with OS. However, no significant differences were found in minor complications (Clavien I-II) and major complications (Clavien III-V). PCNL provided

Fig 5. Forest plot and meta-analysis of blood loss(h) and blood transfusion(I).
https://doi.org/10.1371/journal.pone.0206810.g005 significantly shorter operative times and hospitalization times compared with OS. And PCNL provided significantly less blood loss and blood transfusion compared with OS. In subgroup analysis, there was no significant difference for overall complications and operative times between mini-PCNL and OS. In sensitivity analysis, there was no significant difference for overall complications between PCNL and OS.
In the application of urolithiasis surgery, SFR represents a paramount important parameter. Al-Kohlany KM et al. [7] also found that both PCNL and OS were comparable in regard to SFRs at discharge home and at follow-up. However, Zhang FBY et al. [6] found that OS provided a significantly higher final SFR compared with PCNL (97.5% vs 76.1, p < 0.001). Our pooled data showed that although OS provided a significantly higher immediate-SFR compared with PCNL, no significant difference was found in final-SFR. And the results of the subgroup and sensitivity analysis of standard PCNL and mini-PCNL were consistent with the overall results.
The safety of the patients is also an important parameter. Our pooled data showed that PCNL provided significantly lower overall complication rate compared with OS. And no significant differences were found in minor complications (Clavien I-II) and major complications (Clavien III-V). However, there was no significant difference for overall complications in subgroup analysis between mini-PCNL and OS. And no significant difference was found for overall complications between PCNL and OS in sensitivity analysis. Our pooled data also found that although PCNL provided significantly shorter operative times compared with OS, no significant difference was found for operative times between mini-PCNL and OS in subgroup analysis. The disadvantage of small instruments is that it is necessary to fragment staghorn stones into smaller pieces that fit through the narrower sheaths, which would increase the operative times of the mini-PCNL. Prolonged operating times lead the trend towards higher complications for mini-PCNL compared with standard PCNL.
A great deal of studies [12][13][14] found that PCNL provided significantly less blood loss compared with OS. While Zhang FBY et al. [6] found that OS provided less blood loss compared with OS, though with no statistically significant. However, our pooled data showed that PCNL provided significantly less blood loss and blood transfusion compared with OS. And the results of the subgroup and sensitivity analysis of standard PCNL and mini-PCNL were consistent with the overall results.
Several limitations of our meta-analysis should be taken into consideration when interpreting the results. First of all, nine of ten studies were non-RCTs. Those results should be interpreted with caution given the potential for selection and treatment bias due to the non-RCT heterogeneous nature. Heterogeneity among studies were found to be high for several parameters, including final SFR, operative times, hospitalization times, and blood loss. Although between-study heterogeneity was significantly reduced by the sensitivity analysis for final SFR and blood loss, heterogeneity of operative times and hospitalization times remained statistically significant. The differences in surgical technique and surgical experience, and outcome definitions have all acted an important role in the heterogeneity. The surgical technique for PCNL and OS, the diameters and location of stones were not similar across the different studies. Overall, only 10 studies with 921 patients could be included in the meta-analysis. Such a small number of studies were unable to make strong conclusion. Thus, further large sample prospective, multi-centric studies and RCTs should be undertaken to confirm our findings.
Nevertheless, our meta-analysis was conducted at an appropriate time. Ten data have accumulated for analysis. We applied multiple strategies to identify studies, strict criteria to include and evaluate the methodological quality of the studies, and subgroup analysis to minimize the heterogeneity. Hence, we provide the most up-to-date information in surgical treatment of patients with staghorn calculi, and this could guide urologists and patients to decide on the surgical method, and to select the optimal therapy.

Conclusion
Our systematic review and meta-analysis demonstrated that standard PCNL seems to be a safe and feasible alternative compared to OS or mini-PCNL for patients with staghorn stones with many advantages, such as shorter hospitalization times and operative times, less blood loss and blood transfusion, and without increasing complications nor decreasing final SFR. However, our conclusion should be treated prudently and further large sample prospective, multi-centric studies and randomized control trials should be undertaken to confirm our findings.