Introduction

Kidney stone disease (KSD) is rising with a lifetime prevalence of 14% [1,2,3]. Surgical options such as shockwave lithotripsy (SWL), percutaneous nephrolithotomy (PCNL), and ureteroscopy (URS) are all used as treatment modalities [4,5,6]. The chosen treatment often depends on stone characteristics, patient fitness, comorbidities, surgical expertise, and underlying renal function. Preoperative assessment of these patients involves up-to-date imaging, urine culture, renal function, and fitness for a general anesthetic. The overall incidence of KSD has been rising, and hence more patients are subjected to surgical intervention [7, 8].

Kidney function can be impaired as a result of the disease, urinary infections, or ureteric obstruction related to the stone or surgical intervention related to the KSD. While it is generally believed that treatment of KSD would lead to an improvement of renal function, it is unclear if the surgical procedure required to remove the stone will have an adverse effect. There is a theoretical risk of deterioration of renal function with both PCNL and URS. The physical puncturing of the kidney during PCNL causes direct damage to the renal parenchyma, and this is amplified as PCNL is increasingly used to treat complex or staghorn calculi requiring multiple puncture tracts [9]. During endoscopic approach to the urinary tract, high pressure irrigation is often required to maintain a visual field, causing dilatation of the renal calyces that could potentially harm the function of the kidney. In addition, while the renal parenchyma is not breached as with PCNL, application of the holmium:yttrium-aluminum-garnet (Ho:YAG) laser may cause heat related tissue damage [10, 11].

Given the theoretical risk of renal function decline with both PCNL and URS, our aim was to conduct systematic review to clarify the effect of endourological interventions on renal function.

Method

Search Strategy

Our systematic review was performed as per the Cochrane guidelines and Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) checklist [12]. The databases searched included MEDLINE, Cumulative Index to Nursing and Allied Health Literature (CINAHL), Excerpta Medica Database (EMBASE), Scopus, Clinicaltrials.gov, Google Scholar, Cochrane library and Web of Science with references cross-checked, and individual urology journals also hand-searched. The search terms included “stones,” “calculi,” “urolithiasis,” “nephrolithiasis,” “kidney,” “renal,” “ureteroscopy,” “URS,” “laser,” “fragmentation,” “percutaneous,” “PCNL,” “mini,” “miniaturized,” “percutaneous nephrolithotomy,” “lithotripsy,” “renal function,” “kidney function,” “chronic kidney disease,” “CKD,” “creatinine,” “eGFR,” “MAG3,”and “DMSA.” The references of identified studies were examined to find any further potential studies for inclusion. Boolean operators (AND, OR) were employed. The research was limited to English language articles from 1990 to June 2019.

A cut off of ten patients was set to include studies from centers with minimum relevant endourological experience in managing stones. All original studies were included, and where more than one article was available, the study with the longest follow-up was included. Experienced reviewers (TR, AP) not involved in the original work independently identified all the studies that appeared to fit the inclusion criteria, which were then included for a full review. All discrepancies were resolved with mutual agreement and consensus with the senior author (BKS).

Inclusion Criteria

  1. 1.

    Studies reporting on renal function of patients following endourological intervention (PCNL and URS)

  2. 2.

    Studies reporting on a minimum of 10 patients

  3. 3.

    Studies available in English

Exclusion Criteria

  1. 1.

    Laboratory, animal data, or review articles

  2. 2.

    Studies published before 2000

Data Extraction and Analysis

The following variables were extracted from included studies: author, year of publications, journal, country of study, treatment modality, patient characteristics, stone characteristics, method of monitoring renal function, follow up, and pre- and postoperative renal function. Data were collected using Microsoft Excel 2019 (version 16.28). Due to the heterogeneity of the included studies, the authors decided that meta-analysis of effect sizes was not suitable, and hence either pooled analysis was performed to calculate mean values or outcomes were summarized in a narrative fashion.

Quality of Studies Assessment

The Centre for Evidence-Based Medicine criteria were used to evaluate the levels of evidence of the included studies [13]. The quality of reporting outcomes was performed according to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement [14].

Results

Study Selection and Characteristics

The literature search yielded 837 publications (Fig. 1). After excluding reports that were out of the scope of our systematic review, 145 abstracts were reviewed of which 39 full articles were reviewed for inclusion. Twenty-eight studies were included in the final review (5 were excluded as they were published before 1990, 4 did not mention the effect on renal function, 1 was an animal study and 1 was not in English language). Included studies were published between 2001 and 2019. Three papers compared the effect of PCNL and URS on renal function (Fig. 1).

Fig. 1
figure 1

PRISMA flow chart of the included studies

Full size image

PCNL

The effect of PCNL on renal function was assessed in 21 studies published between 1999 and 2019. This included 1994 patients, and the mean age of patients was 49.3 years (Tables 1 and 2). The follow-up in these studies ranged from 1 day to 51 months [15, 16]. While 11 studies [15,16,17,18,19,20,21,22,23,24,25,26] used blood test to measure renal function, 1 study [27] used radionucleotide scans and 8 studies used combination of both [28,29,30,31,32,33,34,35].

Table 1 Study details
Full size table
Table 2 Effect of PCNL and URS on the renal function of patients
Full size table

Three studies showed significantly improved renal function following PCNL [19, 30, 35]. Eight studies showed no significant improvement but a trend toward improved renal function [16, 17, 23,24,25, 28, 29, 34]. Eight studies showed no significant change in renal function [18, 21,22,23, 26, 27, 32, 34]. Handa et al. showed that on day 1 post-procedure the renal function was significantly worse [15].

Hegarty et al. showed significantly worse renal function in patients who underwent multiple tracts PCNLs, but no significant change in those with single tract approach [20]. Fayad et al. showed that those with poor preoperative renal function had significantly worsened renal function post-procedure, but those with normal preoperative function had a stable renal function [31]. Fayad, Ozden, and Chi et al. showed that diabetes was associated with poor postoperative renal function [17, 23, 31]. In addition, Fayad et al. and Ozden et al. showed that postoperative UTI was associated with poor postoperative renal function [17, 31]. Perez-Fentes et al. suggested that postoperative complications were associated with more parenchymal damage following PCNL [33].

Ureteroscopy

The effect of ureteroscopy on renal function was assessed in four studies published between 2014 and 2019 [36,37,38,39] (Tables 1 and 2). This included 608 patients, 355 males and 253 females, and the mean age of patients was 54.9 years (Table 1). The follow-up ranged from 4 weeks to 28.1 months [36, 40]. All 4 studies used blood tests (creatinine, eGFR) for renal function monitoring [36,37,38,39].

Yang et al. [36] showed that URS significantly improve postoperative renal function. The other three studies showed no statistically significant change but trend to improvement in postoperative renal function [37,38,39].

Comparative Studies between PCNL and URS

Three studies included both PCNL and URS published between 2016 and 2019 (Tables 1 and 2). This included 262 patients with a mean age of 57.3 (Table 1) [40,41,42]. The follow-up ranged from 60 days to 90 days. Jiao et al. and Cho et al. used blood tests (creatinine, eGFR) to measure renal function while Piao et al. used combination of blood test and radionucleotide scans [40,41,42].

Both Piao et al. and Jiao et al. showed no significant change in renal function but a trend towards improvement [40, 42]. Cho et al. showed that if preoperative renal function was normal and then postoperative renal function was statistically normal, but if the renal function was abnormal, then it had a tendency to deteriorate significantly postoperatively [41].

Quality Assessment

The Centre for Evidence-Based Medicine criteria were used to evaluate the levels of evidence of the included studies and found that 3 studies were level one [18, 28, 30], 11 were level two [16, 21, 26, 27, 29, 31,32,33, 38, 41, 42], and 14 were level three evidence ( [15, 17, 19, 20, 22,23,24,25, 34,35,36,37, 39, 40])(Table 1). In addition, the quality of all studies was assessed for inclusion against the STROBE criteria [14].

Discussion

Meaning of the Study

Here we present the only systematic review on the effect of PCNL and URS on renal function. Our study suggests that overall renal function is not detrimentally affected by endourological intervention, but there are potentially some important predictive factors including preoperative renal function, diabetes, and hypertension, hence patients should be appropriately counseled and followed up.

For patients undergoing PCNL, the results were varied. Handa et al. showed a significantly worse postoperative renal function but their follow up time frame was only 1 day, and this may not have been replicated at subsequent follow up [15]. Gorbachinsky et al., Hegarty et al., and El-Tabey et al. showed that multiple tracts were predictive of significant deterioration in renal function [19, 20, 32]. This is perhaps a reflection of the theoretical risk of parenchymal damage causing a decline in renal function, but this wasn’t replicated across all studies using multiple tracts. Several studies showed that a poor preoperative renal function was predictive of the postoperative function [23, 31, 41]. Additionally, Fayad et al. showed that diabetes and hypertension were independent risk factors for poor outcome [31], El-Tabey et al. showed that postoperative bleeding was a factor [19], and Ozden et al. showed that diabetes and urinary tract infection were independent factors [23]. This suggests that declining renal function maybe attributable to patient comorbidities and other underlying disease as opposed to the effect of the endourological procedure alone. Especially as three studies showed significant improvement in function, and the majority of others showed a trend toward improvement [19, 30, 35].

With patients undergoing ureteroscopy only, Yang et al. showed a significant improvement in postoperative renal function [36]. Cho et al. demonstrated that poor preoperative renal function predicated deterioration, but the renal function was protected for those with good pre-operative renal function [41]. All the other studies showed a trend towards improvement of renal function. Interestingly Sninksy et al. concluded that there was no association between poor preoperative function or multiple procedures on the post-procedural function [37].

Strengths, Limitations, and Areas for Future Research

This study gives and overview of the effect of endourological techniques effect on renal function. Due to the heterogeneity of the studies and methods for monitoring renal function meta-analysis was not possible; this also made it difficult to compare the studies directly. The patient population inherently contains a number of confounders in terms of comorbidities. In addition, many of papers were retrospective case series and prone to bias. It is prudent that future studies look at the procedural cost differences and quality of life in these patients [43,44,45]. Similarly, the laser settings and the heat generated by them need to be addressed especially in the context of patients with poor-preoperative renal function [46].

The review highlights that although the renal function is unaffected in most endourological interventions, yet there is a lack of prospective real-life data addressing this issue. Similarly, perhaps there is a need for a randomized control trial addressing both PCNL and URS, with an emphasis on pre- and postoperative renal function, taking into consideration the comorbidities such as diabetes, hypertension, obesity, and chronic kidney disease [47]. This is especially important as previous studies have shown a direct link of these factors on the renal function [48]. Identification of high-risk patients and periodic monitoring of renal function would help in early intervention and is likely to protect further deterioration [49]. PCNL does not seem to result in loss of renal function [29]. However, increasing multiplicity of tracts seems to negatively impact the renal function [50]. Minimally invasive PCNL however does not seem to effect renal function even when there are multiple tracts [35]. In patients with pre-existing CKD or diabetes/hypertension and non-obstructed pelvicalyceal system multi-tract PCNL may result in a kidney function deterioration and thus endoscopic combined intrarenal surgery (ECIRS) should be contemplated [51].

Conclusion

This review suggests that endourological interventions do not adversely affect renal function and tend to improve it in patients who do not have a poor renal function prior to the procedure. Several factors including poor preoperative renal function, diabetes, hypertension, and multiple percutaneous tracts appear to predispose patients to declining renal function after procedure, and these patients should be counseled for and followed up appropriately.