World J Urol DOI 10.1007/s00345-014-1367-5

Topic Paper

Lessons learned from the CROES percutaneous nephrolithotomy global study Guido M. Kamphuis · Joyce Baard · Matias Westendarp · Jean J. M. C. H. de la Rosette 

Received: 6 February 2014 / Accepted: 18 July 2014 © Springer-Verlag Berlin Heidelberg 2014

Abstract  Purpose  The purpose of the study was to give an overview of the data derived from the CROES PCNL Global Study published previously in 25 articles. Methods A comprehensive overview of the outcome of the CROES PCNL Global Study was made, analysed and compared with the current literature and guidelines. Results Percutaneous nephrolithotomy (PCNL) was predominately performed in prone position. Although the supine position claims to be favourable over the prone approach, the present study showed a longer operation time and lower stone-free rate (SFR). This might be explained by differences in definition in operation time and methods in the evaluation of residual stones. Ultrasound (US)-guided access proves beneficial in lowering puncture time and radiation exposure. Renal anomalies can safely be treated by PCNL and have similar outcomes to a normal situation. In patients with a solitary kidney, however, there is a lower SFR and more bleeding. Also, severe chronic kidney disease (CKD) patients have less favourable outcome. Morbidity and complications following PCNL are dominated by fever (10.5 %) and bleeding (7.8 %). A matched control analysis confirmed that antibiotic prophylaxis gives a threefold lower post-operative fever rate. In a multivariate analysis, it was elegantly demonstrated that bleeding was directly related to the size of the dilatation: the larger the bore, the higher the chance for bleeding. Elderly Guido M. Kamphuis and Joyce Baard equally contributed for this work. G. M. Kamphuis · J. Baard · M. Westendarp · J. J. M. C. H. de la Rosette (*)  Department of Urology G4‑172, Academic Medical Center, University of Amsterdam, Meibergdreef 9, 1105 AZ Amsterdam, The Netherlands e-mail: [email protected]

patients are at higher risk of complications and longer hospital stay. Overall, obese patients have similar outcome as the general population; however, super-obese (BMI > 40) patients have a higher chance of more severe complications. For the first time, this database illustrated a significant relationship between results and complications of PCNL, and caseload volume. The optimal case volume per centre appears to be 120 PCNL’s/year. Finally, a nomogram has been developed that enables better patient counselling and decision-making. Conclusion  The CROES PCNL Global Study is the largest real-life study providing new insights into general and special conditions. Because of the vast number of patients included, rare conditions including renal anomalies, solitary kidneys and patient characteristics like severe CKD, super obesity and old age could be analysed. Besides this information, a nomogram was developed. And for the first time, the influence of caseload volume was established. Keywords Percutaneous nephrolithotomy · Renal stones management · Outcome · Complications Abbreviations PCNL Percutaneous nephrolithotomy CROES Clinical Research Office of the Endourological Society SFR Stone-free rate NT Nephrostomy tube CT Computed tomography ESWL Extracorporeal shock wave lithotripsy URS Ureterorenoscopy AB Antibiotics ASA American Association of Anaesthesiologists HU Hounsfield unit BMI Body mass index CKD Chronic kidney disease

13



Introduction The Endourological Society was established to facilitate scientific dialogue among endourologists worldwide. The Clinical Research Office of the Endourological Society (CROES) was established in 2008 as an entity for organising, structuring and facilitating a global network for research in endourology [1]. The CROES initiated the PCNL Global Study to establish a prospective global database to study the current indications and outcomes of percutaneous nephrolithotomy (PCNL). The purpose of the study was to better understand the fundamental differences between clinical institutions around the world in the use of this procedure and to identify specific factors that might influence results and treatment-related morbidity. The study was open between November 2007 and December 2009. In this manuscript, we present a comprehensive overview of the study and the main outcomes that have been published in 25 articles. We especially highlight the lessons learned from this global study and reflect on its limitations and impact on our daily clinical work and directions for future research.

Patients and methods Study objectives, organisation, inclusion criteria, treatment protocol and follow-up have been described in the first publication [2]. In short, the primary study objective was to assess the current indications for PCNL treatment and outcomes in terms of the stone-free rate (SFR). Secondary study objectives were to assess the perioperative morbidity (30 days) by using the modified Clavien–Dindo grading system and to define risk factors for the development of perioperative morbidity. The authors were invited by the CROES council to form a globally representative Steering Committee. Their objective was to recruit 100 centres worldwide. Each centre was invited to include all patients who were treated consecutively for 1 year, with the study period at each site starting with the treatment of the first included patient. Patients who were eligible for inclusion were all those who were candidates for PCNL without specific exclusion criteria. The procedure was considered completed when all removable stones had been taken out. PCNL positioning, imaging, tract formation technique, exit strategies and post-operative renal assessment were performed according to the judgment of the surgeon. In this paper, we present, in “General results”, an overview of the data that have been published in several articles from the CROES PCNL Global Study. In “Discussion”, we put these results in perspective to a selection of studies and guidelines and critically reflect on the insights this study brought, as well as the questions that have been raised.

13

World J Urol Table 1  General results of main features of global PCNL study Positioning 80 % prone 20 % supine Access 90 % Urologist 10 % Radiologist Puncture location 83 % below 12th rib 15 % above 12th rib 2 % above the 11th rib Dilating technique 41 % balloon dilation 59 % serial or telescopic dilation Fragmentation devices 42 % pneumatic 24 % ultrasonic device 18 % combined (pneumatic/ultrasonic) 7 % laser 1 % electrohydraulic 8 % no device Exit strategy 91.2 % of patients NT SFR 76 % overall Secondary procedure 15 % Complications 85.5 % uneventful procedure 7.8 % bleeding 3.4 % perforation of the renal pelvis 1.8 % hydrothorax 1.7 % failure to complete the procedure 10.5 % fever 5.7 % blood transfusion

General results Between November 2007 and December 2009, 5,803 patients were treated at 96 centres globally: Europe (3,249), Asia (1,369), North America (830), South America (345) and Australia (10). The first publication describes the first analysis of this large database [2]. It taught us the general aspects of the PCNL procedure and general outcome globally. A summary of the main features is found in Table 1. Surgery and access AB prophylaxis/positioning/access technique/tract dilation/ lithotrite technique A group of 162 patients in the database with negative pre-operative urine culture who did not receive antibiotic (AB) prophylaxis were matched (on pre-operative nephrostomy tube, staghorn calculi and diabetes status) with an equal number of patients who did receive AB prophylaxis.

World J Urol

Patients in AB prophylaxis group were 5.2 years younger. Patients who received AB prophylaxis were less likely to experience fever (2.5 vs. 7.4 % p = 0.04) and other postoperative complications (1.9 vs. 22 %) and had a higher SFR (86.3 vs. 74.4 %) [3]. In this study, the vast majority (80.3 %) of patients were operated in prone position. PCNL performed in prone position had a shorter median operation time and a higher SFR (77 vs. 70 % in patients in supine position). The blood transfusion rate of patients operated in prone position however was higher (6.1 vs. 4.3 %). The chance of a failed procedure was higher in the supine positioned group (2.7 vs. 1.5 %). The question that remains is the influence of patient selection on these results. Patients in prone positions were significantly younger, more males and had a lower average ASA score. Also, the use of anticoagulants, prednisone and stone burden was in favour of the prone positioned group, although not significantly [4]. The effect of prone or supine technique in PCNL for patients with a staghorn stone was separately analysed. Comparing 232 patients with a staghorn stone who were treated in supine position, with 1,079 patients with a staghorn stone treated in prone position, there was no difference in complication rate found. The SFR in the prone position operated patients was higher with 59.2 % compared to 48.4 % in supine position [5]. A matched analysis was done to evaluate the effect of imaging modality used for percutaneous renal access. On univariate analysis, fluoroscopy was related to a higher incidence of haemorrhage. Multivariate analysis however revealed the relationship of haemorrhage with greater size access sheath and multiple punctures. Failed access was significantly more frequent in the fluoroscopy group (2.9 %) compared to the US group (0.7 %) [6]. The patients in the database who had an isolated upper pole calyx access (n = 403) were compared to the patients who had an isolated middle calyx access (n  = 717) and lower pole calyx access (n  = 3,112). Patients with multiple accesses were excluded. Patients with an isolated upper pole calyx access had a greater stone burden, and there were more patients with staghorn stones in this group compared to the other access groups. Operation time was longer (92 vs. 75 min), SFR lower (77.1 vs. 81.6 %) and complication rate higher (23.5 vs. 16.1 %) in the isolated upper pole calyx access group. Isolated upper pole access was concluded to be indicated in a select group of patients and should be seen as a higher complex procedure compared to lower pole access [7]. Different ways of tract dilatation were compared showing global differences in tract dilations with a preference for telescopic/serial dilation in Asia and South America and a predominant use of balloon dilation in North America. In Europe, both dilation systems were used equally. Operation

time was significantly longer in the balloon dilation group with a mean of 94 min compared to 60 min in the serial dilation group. We extensively searched the database, but could not identify a logical explanation. The risk of bleeding complications was significantly higher in the balloon dilation group with 9.4 % compared to the telescopic/serial dilation group with 6.7 %. Also, the blood transfusion rate was significantly higher in the balloon dilation group (7 vs. 4.9 %) [8]. Multivariate analysis, however, showed that sheath size and not dilation method was the predictive of bleeding complications [9]. A small sheath size of 18 F or less was used in 271 patients with 1.1 % blood transfusions. 1,039 patients had a 24–26 F sheath with a blood transfusion rate of 4.8 %. 3,533 patients were treated with a large sheath defined as 27 till 30 F and had a blood transfusion rate of 5.9 %, and 371 patients were treated with a 32–34 F sheath with a blood transfusion rate of 12.1 %. Other predictive factors of bleeding complications distinguished were operating time, stone load and caseload volume. BMI and anticoagulation therapy were not associated with an increased risk of bleeding complications. The effect of type of lithotrite technique on post-operative fever was studied. The data of patients treated with pneumatic fragmentation, ultrasonic or a combination of these two, laser fragmentation and no fragmentation, were compared. After adjustment for confounding factors, no difference in post-operative fever was found [10].

Kidney and stone characteristics Renal anomalies/solitary kidneys/stone type A major strength of the CROES PCNL Global Study is the large number of patients included. A rare condition like renal anomaly was found in 202 included patients (3.6 %) out of 5,542 in whom renal anatomy was recorded [11]. Most frequent anomalies were horseshoe kidneys in 1.8 % and malrotated kidneys in 1.3 %. SFR and complication rate were not different in PCNL between patients with ‘normal’ kidneys and renal anomalies. Median operative time was significantly longer (87 vs. 75 min), and access for PCNL was unsuccessful in more patients with a renal anomaly (5 vs. 1.7 %). In 5,745 out of 5,803 patients, there was information on solitary or bilateral kidneys. 189 patients (3.3 %) had a solitary kidney. Comparison of patient characteristics showed that the patient groups were in general comparable, apart from higher ASA scores, higher cardiovascular comorbidity, higher chance of previous kidney stone treatment and renal impairment in the solitary kidney group. Bleeding rates were comparable, but transfusion rates were higher in

13



the solitary kidney group. SFR was significantly lower in the solitary kidney group (65 vs. 76 %) [12]. Separate evaluation of single large non-staghorn renal stones in the CROES PCNL Global Study found that the location of the stone in a calyx was associated with decreased fitness for surgery, an increased risk of postoperative complications compared to renal site. Not surprisingly, an increase in stone size resulted in a lower SFR and higher rates of post-operative fever and blood transfusion [13]. Evaluation of the impact of Hounsfield units (HU) of renal stones was possible in 1,011 patients in whom data of HU of renal stones were collected. The patients were divided into two groups: high HU > 1,000 (n  = 449) and low HU ≤ 1,000 (n  = 562). Blood transfusion rate was with 6.5 % in the low HU group, almost double the blood transfusion rate in the high HU group with 3.3 %. Operation time and re-treatment rate were comparable. Another interesting finding was that the optimal stone density for SFR was 1,250 HU. Furthermore, in both groups, it was noticed that the more extreme the HU of the stones, the longer the operation time. In conclusion, very low and high stone densities were associated with lower rates of treatment success and longer operating time [14].

Patient characteristics Age/BMI/CKD With the increasing life expectancy, surgeons are more often confronted with elderly patients for surgery. The patients included in the database were divided into three age groups: below 18 years, between 18 and 70 years, and 70 years and older [15]. The 70 years and older patient group was compared to the group aged 18 till 70 years. There were significantly more staghorn stones in the older patient group, but mean stone size was comparable. Comorbidity rate was higher. SFRs were similar. In matched analyses, a statistical higher rate of complications was found in the older patient group. Also, the hospitalisation time was higher (5 vs. 4.1 days) in the elderly group. To evaluate the effect of BMI on the success and complications of PCNL, the database was used [16]. Patients with solitary kidneys, a history of PCNL and abnormal renal anatomy were excluded. 3,709 patients were selected this way. 62 % had a BMI 25 or more. Patients were divided into normal weight (BMI 18.5–25, n = 1,394), overweight (BMI 25–30, n  = 1,568), obese (BMI 30–40, n  = 650) and super obese (BMI > 40, n  = 97). Hospital stay was independent on BMI. There was no difference in overall complication rate found. However, a higher rate of major

13

World J Urol

complications was found in the super-obese patients compared to normal-weight patient group (10.5 vs. 3.7 %). SFR was 77.5 % in the normal-weight group, 79.7 % in the overweight group, 78.9 % in the obese and 65.6 % in the super-obese group (p = 0.009). All patients in the database were divided according to kidney function into three groups: chronic kidney disease (CKD) class O/I/II—eGFR > 60 ml/min, CKD stage III— eGFR 30–59 ml/min and CKD stage IV–V—eGFR 70 years old is independent risk factor for complications and longer hospital stay BMI BMI >40: higher rate of major complications CKD

Patients with CKD IV–V: lower SFR, longer operation time, longer hospital stay and higher complication rate

Results

Nomogram developed using only clinical and radiological parameters

were a minor complication. Two patients died in the postoperative period because of urosepsis. In multivariate analysis as risk factors for complications, the use of anticoagulants, positive urine culture, cardiovascular comorbidity and longer operation time (>74 en >115 min) was found [20]. About 10 % of patients developed fever after PCNL despite AB prophylaxis. The chance of fever post-PCNL increased in the presence of positive urine culture, diabetes, staghorn calculi and a pre-operative nephrostomy tube [21]. As a spin-off, the investigators of the PCNL Global Study assessed the inter-rater agreement for Clavien–Dindo scores of the complications of PCNL. Especially for minor complications, the inter-rater reliability proved to be low. To improve the reliability and consistency of reporting adverse outcomes of PCNL in the future, a categorisation of PCNL complications according to Clavien–Dindo classification has been proposed [22]. The impact of case volume, after adjustment for possible confounders on outcome, was studied in the database. SFRs were significantly higher in high-volume centres (>77 cases/year). Multivariate analysis showed that SFR improved with increasing caseload with a peak at 120 cases per year. Apart from higher SFR, patients in high-volume centres had fewer minor and major complications compared to those in low-volume centres. A nadir of 120 cases per year was found. Also, the hospital duration was shorter in the high-volume case group [23]. Using the database, Smith et al. [24] developed a nomogram to predict SFR using preclinical and radiological

Advantage of AB on outcome and complications rate

Safe procedure Ultrasound guidance beneficial No difference in SFR and complication rate

PCNL technically feasible and safe in obese patients; in super-obese patients, the SFR is lower

information. Predictions based on the nomogram are clinically useful if the threshold of 60 % estimated SFR or more is used to determine the treatment strategy.

Discussion The CROES was a success in terms of the organisation of a large-scale, global research where prospectively information was collected on patients treated with PCNL. The scale of the research made the database a feast for data managers to analyse. It has been successful in giving information on PCNL treatment, outcome and complications in general population and in subpopulations with rare conditions like renal anomaly or BMI over 40. Also, the possibilities of ‘matched analysis’, because of the large numbers, enabled it to produce high-level evidence, although it was not a randomised trial. The advantage of AB prophylaxis in uncomplicated patients with negative pre-operative urine culture is a clear example of information that would be, almost, impossible to obtain in another way. Looking back, the absence of information on stone formation and clear definition on certain concepts like ‘operation time’ and ‘stone-free’ are drawbacks to be kept in mind when analysing the data and interpreting the publications. In this discussion, it is impossible to go over all the discussions that have been described in the 25 articles on the CROES PCNL Global Study. We try to present a comprehensive overview of the major points of the discussion.

13



Overall SFR achieved found in the CROES PCNL database was 76 %. Fifteen per cent of patients underwent a secondary procedure. Several studies have found a higher SFR [25]. Even more, it has to be kept in mind that SFR was defined as the absence of fragments bigger than 4 mm on KUB, US or CT. Therefore, it is likely that the overall SFR is lower than reported. One of the possible explanations is that in the CROES database, not only expert stone centres collected their data, but also the every day urology practice. Antibiotics Different studies have been conducted assessing antibiotics in PCNL, almost all focusing on the type of antibiotic, the dosage and the duration. Overall, no significant difference is seen in preventing septic complications between a single (prophylactic) dose of AB and an additional treatment [26– 28]. The EAU guideline in infections as well as the AUA guideline in staghorn calculi advice prophylactic use of AB in PCNL; however, a specific regimen is not described [29, 30]. As mentioned in several reviews, there are few studies defining the risk of post-operative infection after endoscopic procedures and no clear-cut evidence exists [31, 32]. Only one RCT is performed comparing placebo with AB prophylaxis in patients undergoing an endoscopic stone surgery. This multicentre, placebo-controlled, double-blind study investigated the efficacy of a single dose of cefotaxime 1 g i.v. in the prophylaxis of patients undergoing endoscopic extraction of urinary tract stones. No significant difference was seen in the incidence of post-operative fever. There was a significant higher incidence of post-operative bacteriuria between the first and third post-operative day in the placebo group (15/60, 25 %) than in the cefotaxime group (5/60, 8.5 %) [33]. Bootsma et al. [31] state in their systematic review that there is low evidence suggesting a better outcome after PCNL when prophylactic AB is given in patients in whom pre-operative urine cultures are negative. In contrary, analysis of the CROES PCNL Global Study showed a clear advantage of AB prophylaxis in uncomplicated patients with negative pre-operative urine cultures on outcome and complications rate. Because of ethical concerns with a controlled trial, randomising patients for AB prophylaxis versus placebo will not be easy to conduct today. This matched group analysis provides the second best evidence for the principle of AB prophylaxis in PCNL. Positioning There is an ongoing discussion about positioning of the patient for PCNL in prone or supine position [34]. As the CROES study shows, the majority of urologists globally

13

World J Urol

favour prone position. PCNL seems to be as safe in supine as in prone position, with comparable complication rates. Only two RCTs are conducted on this subject in which equal overall results in terms of efficacy and safety were seen [35, 36]. Compared to the CROES study, two contradictory outcomes are found. First, a significant higher operation time was observed for the supine position, and second a significantly higher stone-free rate in the prone position PCNL group was seen. These contradictory outcomes might be explained by the differences in the definition of operation time and different methods and protocols for assessing a stone-free status. There was no clear definition on operation time. Operation time could start at first patient positioning, or at skin puncture for access, making it impossible to correctly interpret this parameter. Because of possible advantages of supine position in several specific patient comorbidity, as an endourologist, it is of importance to have experience with both prone as supine positions. Ultimately, the decision on how to position the patient during PCNL should be made on patient’s characteristics and surgeons’ preference. Imaging modality A prospective randomised trial was conducted by Agarwal et al. to evaluate the safety and efficacy of US-guided renal access in PCNL compared with conventional fluoroscopy-only-guided renal access. The mean numbers of attempts for successful puncture were higher in the fluoroscopy-only-guided group, and the mean time to successful puncture was shorter [37]. A comparable randomised trial was conducted by Basiri et al. [38] who also found a significant shorter duration time of access in the US-guided group. The CROES database does not mention data about access time of the US-guided and fluoroscopy group but a significantly shorter PCNL procedure was found in the US-guided group compared to the fluoroscopy group (79.3 vs. 84.6 min; p = 0.039). Failed access was not commonly seen, but significantly more frequent in the fluoroscopy group compared to the US group. Tract dilation In general, metallic telescope dilators, Amplatz fascial dilators, or a balloon dilator can achieve dilation of the percutaneous access tract. The first two options are less expensive but are more time-consuming and require longer X-ray exposure when fluoroscopy is used. Further, no difference in morbidity is seen in the different dilation techniques [39]. As discussed in “General results”, the risk of bleeding complications was significantly higher in the balloon dilation group with 9.4 % compared to the telescopic/serial

World J Urol

dilation group with 6.7 %. Also, the blood transfusion rate was significantly higher in the balloon dilation group (7 vs. 4.9 %) [8]. There are several papers that conclude that the gradual balloon dilation is less harmful and at risk for bleeding than the Alken dilation [40–42]. The contradictory in the CROES database results consequently fuelled a multivariate analysis identifying sheath size as the most plausible cause for this [9]. In the CROES study, a longer operative time was seen in the balloon dilation group. This remains somewhat obscure and may be explained by parameters not captured in the database. For example, one could hypothesise that balloon dilation may be more frequently used in training centres and consequently be related to longer operation times. Renal anomalies Because of the vast number of patients included, sub-analyses on treatment outcomes in rare conditions like renal anomalies could be made. There are few articles dedicated to this subject in the literature to compare the results with, as it is difficult to collect large series on rare conditions. In the CROES study, SFR and complication rate were not different in PCNL between patients with ‘normal’ kidneys and renal anomalies (76.2 vs. 73.6 %). Comparable SFRs are reported in the review of Ganpule and Desai [43]. Median operative time was significantly longer, and access for PCNL was unsuccessful in more patients with a renal anomaly as can be explained by the uncommon anatomy. In the CROES database, 189 patients had a solitary kidney. Compared to patients with two functioning kidneys, patients with a solitary kidney had significantly lower SFR. Basiri et al. [44] evaluated the safety and efficacy of PCNL for the management of large renal stones between 30 patients with single functioning kidneys and 30 patients with double kidneys and found no difference in terms of success rate or complications. Patient characteristics The CROES PCNL Global Study found a higher complication rate in the older patient group and old age (70 years and over) to be an independent risk factor for complications and longer hospital stay. SFR was similar in both age groups [15]. In contrast, when comparing the young versus elderly patients (>70 years) in a series of 673 patients, Anagnostou et al. [45] showed no significant differences in SFR and complications, although a trend in favour of the younger group was seen. In a series of 2,954 patients with renal stones, AbdelKahlek et al. [46] showed a significantly worse stone-free rate after ESWL in patients older than 40 years old. The outcomes of ureteroscopy in elderly population are in

general similar to that of the adult patients. Akman et al. [47] stated that URS is safe and effective for geriatric patients with moderate kidney stones, with a similar and comparable overall SFR after a single-stage procedure compared with PCNL (82.1 vs. 92.8 %). Fuller et al. showed on behalf of the CROES the effect of BMI on the success and complications of PCNL. There was no difference found in overall complication rate, but a higher rate of major complications was found in the subcategory of super-obese (BMI >40) patients compared to normal-weight patients (10.5 vs. 3.7 %, p 0.009) [16]. In a study conducted by Alyami et al. [48], no differences were seen in the different BMI groups, even in the super-obese group. This may be explained by the difference in group size between the two study populations. Body mass index (BMI) has proven to be no risk factor for complications or outcome for stone treatment by URS [49]. Furthermore, we know that BMI is a factor in success of stone treatment by ESWL [50]. When treating superobese patients, the choice between PCNL, ESWL and URS should be influenced by these facts. Chronic kidney disease Multiple and consistent data directly relate the presence of kidney stones to the deterioration of kidney function [46, 47, 51, 52]. Clearly, a complex interaction among stone disease, interventions, chronic infection and poor drainage has its impact on renal function. The CROES PCNL Global Study found a relationship of lower SFR, longer operation time, longer hospital stay and higher complication rate in patients with CKD IV–V. Because kidney stones are a potentially preventable cause of kidney failure, vigilance and close monitoring of CKD patients for nephrolithiasis is important. Bore size of nephrostomy tube and exit strategy After Bellman et al. [53] demonstrate tubeless PCNL decreased post-operative pain and hospital stay without an increase in complications, a still ongoing discussion started. Another, well-designed RCT was performed showing good results of totally tubeless (TTL) PCNL [54]. Interestingly, the CROES PCNL Global Study found that 91.2 % of patients received a nephrostomy tube after PCNL [2]. The CROES PNCL Global Study database contains 244 ureter stent-only patients and 68 TTL patients. Cormio et al. matched these patients to groups with a NT to compare outcomes. The only significant difference was longer operating time and hospital stay in the NT group compared to ST group [18]. In a recent survey presented by Sivalingam et al. [55] among members of the Endourological Society, the

13



majority of the responders (76 %) choose a NT as a postoperative drainage. In conflict with the data from the CROES and this recent survey, the European Association guidelines recommend a TL/TTL PCNL (GR A) in an uncomplicated procedure [56]. On a further analysis on the CROES databases, patients with a nephrostomy tube at the end of the procedure were divided into a small-bore (18 F and less) nephrostomy tube group (SB) and large-bore (>18 F) nephrostomy tube group (LB). There was a significant higher complication rate and a higher drop in haemoglobin in the SB group [19]. This led to the advice that if a nephrostomy tube should be placed at the end of the procedure, a LB NT should be used. This conclusion needs to be put carefully in perspective of bleeding causes that can occur during the whole procedure. Obviously, the quality of the access and patient characteristics is crucial in relation to bleeding complications during and after PCNL. Many of these factors that influence this may not have been captured in the database. Complications Fever and bleeding, urinary leakage, and problems related to residual stones are the most common post-operative complications associated with PNCL [56]. A recent review on complications following PNCL used the validated Dindo-modified Clavien system and showed a normal (uncomplicated) post-operative course in 76.7 % of patients [56]. Other well-designed study showed a complication rate after PCNL of 14 % [57]. A systematic review by Michel et al. found a complication rate of PCNL up to 83 %, most of which were insignificant complications such as bleeding (8 % of which were significant) or fever. However, a transfusion rate is reported as 5–18 %. Major complications such as septicaemia or renal haemorrhage were reported in 0.9–4.7 and 0.6–1.4 %, respectively. Perioperative mortality was rated 0.3–0.78 % [58]. The CROES database showed a rate of complications in accordance with the literature, with an overall complication rate of 15 %. Fever and bleeding were the most usual PCNL complications. The majority were classified as Clavien I. Two deaths were reported, caused by urosepsis. The most important factors associated with higher Clavien complications scores are as follows: ASA score, use of anticoagulant medication, positive microbiologic urine culture, and the presence of concurrent cardiovascular disease. It is surprising that data on complication vary so much. This might be due the way data were collected. It is possible that urologists under-report low-grade complications. The categorisation of post-operative complications using the Clavien–Dindo method has been recommended by the

13

World J Urol

EAU Urolithiasis Panel of experts and recently validated for the PCNL complications [22]. Caseload volume Several studies have been conducted demonstrating the relationship between case volumes and outcomes in surgery [56, 57, 59, 60]. Opondo et al. [23] published on behalf of the CROES about the relationship between caseload volume and SFR and complications in PCNL. Kadlec et al. [61] published a recent series of 7,785 PCNL on the relationship of caseload volume in PCNL and complications. In comparison with the CROES series, they included all centres performing PCNL, even those who performed less than 10 cases per year. These authors found the highest complication rates in the very low-volume centres (1–5 cases/year) and the very high-volume centres (>33 cases/year). Opondo et al. found a decrease in complication rate until a volume of 120 cases per year. After this nadir, the complication rate increased again. They also showed that centres performing large numbers of PCNL achieve higher SFRs. Because of the study design, Kadlec et al. could not assess this subject. Nomogram In a patient with kidney stones, it is important to discuss several possible treatment modalities. A fair comparison of expectations of different treatment modalities is crucial to make a good choice. In 2011, Thomas and colleagues proposed with the Guy’s stone score nomogram, the first nomogram for predicting success of PCNL, based on clinical parameters, literature and expert opinion [62]. The nomogram developed by Smith et al. is based on objective data and only clinical and radiological parameters. Comparison of the two nomograms showed a more accurate predication by the CROES nomogram compared to the Guy’s stone score nomogram [24]. Disadvantage of the nomogram is the use of the definition of ‘stone-free’ as 4 mm or less rest fragments on US, KUB or CT scan. This should be kept in mind when using the nomogram in counselling patients. At the time of the analysis of the CROES database, we faced several questions that could not be addressed since several topics were either not captured in the database or newer developments have taken place in the meantime. For example, data on stone analysis and metabolic analysis were not available and also not weather combined treatment by PCNL and retrograde intrarenal surgery was performed. Moreover, data on the recently popularised mini-PCNL procedures were not captured. Also, the many questions arose regarding the terminology used such as on

World J Urol

SFR. In follow-up, the CROES council has taken the initiative to prepare a systematic analysis in combination with a survey to prepare a document on standardisation of terminology of PCNL [63]. Surprisingly, we have communicated in many papers about indications and outcomes of treatment, whereas we often did not speak the same language. Finally, it is reassuring that the CROES PCNL Global Study has contributed to new insights into stone management by PCNL and encouraged many teams around the world to embark on studies fuelled by the PCNL study.

Conclusion The CROES PCNL Global Study is the largest real-life study providing new insights into general and special conditions (Table 2). Because of the vast number of patients included, rare conditions including renal anomalies, solitary kidneys and patient characteristics like severe CKD, super obesity and old age could be analysed. Besides this information, a nomogram was developed. And for the first time, the influence of caseload volume was established.

References 1. de la Rosette JJMCH (2009) A platform for global endourological research. J Endourol 23(10):1551–1553 2. de la Rosette JJMCH, Assimos D, Desai M, Gutierrez J, Lingeman J, Scarpa R, Tefekli A (2011) The Clinical Research Office of the Endourological Society percutaneous nephrolithotomy global study. J Endourol 25(1):11–17 3. Gravas S, Montanari E, Geavlete P, Onal B, Skolarikos A, Pearle M, Sun YH, de la Rosette JJMCH (2012) Postoperative infection rates in low risk patients undergoing percutaneous nephrolithotomy with and without antibiotic prophylaxis: a matched case control study. J Urol 188(3):843–847 4. Valdivia JG, Scarpa RM, Duvdevani M, Gross AJ, Nadler RB, Nutahara K, de la Rosette JJMCH (2011) Supine versus prone position during percutaneous nephrolithotomy: a report from the Clinical Research Office of the Endourological Society percutaneous nephrolithotomy global study. J Endourol 25(10):1619–1625 5. Astroza G, Lipkin M, Neisius A, Preminger G, De Sio M, Sodha H, Saussine C, de la Rosette J (2013) Effect of supine vs prone position on outcomes of percutaneous nephrolithotomy in staghorn calculi: results from the clinical research office of the endourology society study. Urology 82(6):1240–1245 6. Andonian S, Scoffone CM, Louie MK, Gross AJ, Grabe M, Daels FPJ, Shah HN, de la Rosette JJMCH (2013) Does imaging modality used for percutaneous renal access make a difference? A matched case analysis. J Endourol 27(1):24–28 7. Tefekli A, Esen T, Olbert PJ, Tolley D, Nadler RB, Sun Y-H, Duvdevani M, de la Rosette JJMCH (2013) Isolated upper pole access in percutaneous nephrolithotomy: a large-scale analysis from the CROES percutaneous nephrolithotomy global study. J Urol 189(2):568–573

8. Lopes T, Sangam K, Alken P, Silva B, Saussine C, Shi L, de la Rosette JJMCH (2011) The Clinical Research Office of the Endourological Society percutaneous nephrolithotomy global study: tract dilation comparisons in 5537 patients. J Endourol 25(5):755–762 9. Yamaguchi A, Skolarikos A, Buchholz N-PN, Chomón GB, Grasso M, Saba P, Nakada S, de la Rosette J (2011) Operating times and bleeding complications in percutaneous nephrolithotomy: a comparison of tract dilation methods in 5,537 patients in the Clinical Research Office of the Endourological Society percutaneous nephrolithotomy global study. J Endourol 25(6):933–939 10. Chu DI, Lipkin ME, Wang AJ, Ferrandino MN, Preminger GM, Kijvikai K, Gupta NP, Melekos MD, de la Rosette JJMCH (2013) Lithotrites and postoperative fever: does lithotrite type matter? Results from the Clinical Research Office of the Endourological Society percutaneous nephrolithotomy global study. Urol Int 91(3):340–344 11. Osther PJ, Razvi H, Liatsikos E, Averch T, Crisci A, Garcia JL, Mandal A, de la Rosette JJMCH (2011) Percutaneous nephrolithotomy among patients with renal anomalies: patient characteristics and outcomes; a subgroup analysis of the Clinical Research Office of the Endourological Society global percutaneous nephrolithotomy study. J Endourol 25(10):1627–1632 12. Bucuras V, Gopalakrishnam G, Wolf JS, Sun Y, Bianchi G, Erdogru T, de la Rosette J (2012) The Clinical Research Office of the Endourological Society percutaneous nephrolithotomy global study: nephrolithotomy in 189 patients with solitary kidneys. J Endourol 26(4):336–341 13. Xue W, Pacik D, Boellaard W, Breda A, Botoca M, Rassweiler J, Van Cleynenbreugel B, de la Rosette J (2012) Management of single large nonstaghorn renal stones in the CROES PCNL global study. J Urol 187(4):1293–1297 14. Anastasiadis A, Onal B, Modi P, Turna B, Duvdevani M, Timoney A, Wolf JS, de la Rosette JJMC (2013) Impact of stone density on outcomes in percutaneous nephrolithotomy (PCNL): an analysis of the Clinical Research Office of the Endourological Society (CROES) PCNL global study database. Scand J Urol 47(6):509–514 15. Okeke Z, Smith AD, Labate G, D’Addessi A, Venkatesh R, Assimos D, Strijbos WEM, de la Rosette JJMCH (2012) Prospective comparison of outcomes of percutaneous nephrolithotomy in elderly patients versus younger patients. J Endourol 26(8):996–1001 16. Fuller A, Razvi H, Denstedt JD, Nott L, Pearle M, Cauda F, Bolton D, Celia A, de la Rosette J (2012) The CROES percutaneous nephrolithotomy global study: the influence of body mass index on outcome. J Urol 188(1):138–144 17. Sairam K, Scoffone CM, Alken P, Turna B, Sodha HS, Rioja J, Wolf JS, de la Rosette JJMCH (2012) Percutaneous nephrolithotomy and chronic kidney disease: results from the CROES PCNL global study. J Urol 188(4):1195–1200 18. Cormio L, Gonzalez GI, Tolley D, Sofer M, Muslumanoglu A, Klingler H-C, Stolzenburg J-U, de la Rosette JJMCH (2013) Exit strategies following percutaneous nephrolithotomy (PCNL): a comparison of surgical outcomes in the Clinical Research Office of the Endourological Society (CROES) PCNL global study. World J Urol 31(5):1239–1244 19. Cormio L, Preminger G, Saussine C, Buchholz NP, Zhang X, Walfridsson H, Gross AJ, de la Rosette J (2013) Nephrostomy in percutaneous nephrolithotomy (PCNL): does nephrostomy tube size matter? Results from the global PCNL study from the clinical research office endourology society. World J Urol 31(6):1563–1568 20. Labate G, Modi P, Timoney A, Cormio L, Zhang X, Louie M, Grabe M, de la Rosette JJMCH (2011) The percutaneous nephrolithotomy global study: classification of complications. J Endourol 25(8):1275–1280

13

21. Gutierrez J, Smith A, Geavlete P, Shah H, Kural AR, de Sio M, Amón Sesmero JH, Hoznek A, de la Rosette J (2013) Urinary tract infections and post-operative fever in percutaneous nephrolithotomy. World J Urol 31(5):1135–1140 22. de la Rosette JJMCH, Opondo D, Daels FPJ, Giusti G, Kandasami SV, Wolf JS, Grabe M, Gravas S (2012) Categorisation of complications and validation of the Clavien score for percutaneous nephrolithotomy. Eur Urol 62:246–255 23. Opondo D, Tefekli A, Esen T, Labate G, Sangam K, De Lisa A, Shah H, de la Rosette JJMCH (2012) Impact of case volumes on the outcomes of percutaneous nephrolithotomy. Eur Urol 62(6):1181–1187 24. Smith A, Averch TD, Shahrour K, Opondo D, Daels FPJ, Labate G, Turna B, de la Rosette JJMCH (2013) A nephrolithometric nomogram to predict treatment success of percutaneous nephrolithotomy. J Urol 190(1):149–156 25. Skolarikos A, Alivizatos G, de la Rosette JJMCH (2005) Percutaneous nephrolithotomy and its legacy. Eur Urol 47(1):22–28 26. Tuzel E, Aktepe OC, Akdogan B (2013) Prospective comparative study of two protocols of antibiotic prophylaxis in percutaneous nephrolithotomy. J Endourol 27(2):172–176 27. Seyrek M, Binbay M, Yuruk E, Akman T, Aslan R, Yazici O, Berberoglu Y, Muslumanoglu AY (2012) Perioperative prophylaxis for percutaneous nephrolithotomy: randomized study concerning the drug and dosage. J Endourol 26(11):1431–1436 28. Demirtas A, Yildirim YE, Sofikerim M, Kaya EG, Akinsal EC, Tombul ST, Ekmekcioglu O, Gulmez I (2012) Comparison of infection and urosepsis rates of ciprofloxacin and ceftriaxone prophylaxis before percutaneous nephrolithotomy: a prospective and randomised study. Sci World J 2012:916381 29. Grabe M, Botto H, Çek M, Naber KG, Pickard RS, Tenke P, Wagenlehner F, Wullt B (2013) Guidelines on urological infections. http://www.uromeb.org 30. Preminger GM, Assimos DG, Lingeman JE, Nakada SY, Pearle MS, Wolf JS (2009) AUA guideline on the management of staghorn calculi. http://www.auanet.org 31. Bootsma MJ, Laguna Pes MP, Geerlings SE, Goossens A (2008) Antibiotic prophylaxis in urologic procedures: a systematic review. Eur Urol 54(6):1270–1286 32. DasGupta R, Grabe M (2009) Preoperative antibiotics before endourologic surgery. J Endourol 23(10):1567–1570 33. Fourcade RO (1990) Antibiotic prophylaxis with cefotaxime in endoscopic extraction of upper urinary tract stones: a randomized study. J Antimicrob Chemother 26(suppl A):77–83 34. Baard J, Kamphuis GM, Westendarp M, de la Rosette JJMCH (2014) How well tolerated is supine percutaneous nephrolithotomy? Curr Opin Urol 24(2):184–188 35. Falahatkar S, Moghaddam AA, Salehi M, Nikpour S, Esmaili F, Khaki N (2008) Complete supine percutaneous nephrolithotripsy comparison with the prone standard technique. J Endourol 22(11):2513–2517 36. De Sio M, Autorino R, Quarto G, Calabrò F, Damiano R, Giugliano F, Mordente S, D’Armiento M (2008) Modified supine versus prone position in percutaneous nephrolithotomy for renal stones treatable with a single percutaneous access: a prospective randomized trial. Eur Urol 54(1):196–202 37. Agarwal M, Agrawal MS, Jaiswal A, Kumar D, Yadav H, Lavania P (2011) Safety and efficacy of ultrasonography as an adjunct to fluoroscopy for renal access in percutaneous nephrolithotomy (PCNL). BJU Int 108(8):1346–1349 38. Basiri A, Ziaee AM, Kianian HR, Mehrabi S, Karami H, Moghaddam SMH (2008) Ultrasonographic versus fluoroscopic access for percutaneous nephrolithotomy: a randomized clinical trial. J Endourol 22(2):281–284 39. Wezel F, Mamoulakis C, Rioja J, Michel MS, de la Rosette JJMCH, Alken P (2009) Two contemporary series of

13

World J Urol percutaneous tract dilation for percutaneous nephrolithotomy. J Endourol 23(10):1655–1661 40. Turna B, Nazli O, Demiryoguran S, Mammadov R, Cal C (2007) Percutaneous nephrolithotomy: variables that influence hemorrhage. Urology 69(4):603–607 41. Tomaszewski JJ, Smaldone MC, Schuster T, Jackman SV, Averch TD (2010) Factors affecting blood loss during percutaneous nephrolithotomy using balloon dilation in a large contemporary series. J Endourol 24(2):207–211 42. Davidoff R, Bellman GC (1997) Influence of technique of percutaneous tract creation on incidence of renal hemorrhage. J Urol 157(4):1229–1231 43. Ganpule AP, Desai MR (2011) Urolithiasis in kidneys with abnormal lie, rotation or form. Curr Opin Urol 21(2):145–153 44. Basiri A, Shabaninia S, Mir A, Soltani MH (2012) The safety and efficacy of percutaneous nephrolithotomy for management of large renal stones in single- versus double-functioning kidney patients. J Endourol 26(3):235–238 45. Anagnostou T, Thompson T, Ng C-F, Moussa S, Smith G, Tolley DA (2008) Safety and outcome of percutaneous nephrolithotomy in the elderly: retrospective comparison to a younger patient group. J Endourol 22(9):2139–2145 46. Abdel-Khalek M, Sheir KZ, Mokhtar AA, Eraky I, Kenawy M, Bazeed M (2004) Prediction of success rate after extracorporeal shock-wave lithotripsy of renal stones—a multivariate analysis model. Scand J Urol Nephrol 38(2):161–167 47. Akman T, Binbay M, Ugurlu M, Kaba M, Akcay M, Yazici O, Ozgor F, Muslumanoglu AY (2012) Outcomes of retrograde intrarenal surgery compared with percutaneous nephrolithotomy in elderly patients with moderate-size kidney stones: a matchedpair analysis. J Endourol 26(6):625–629 48. Alyami F, Skinner T, Norman RW (2012) Impact of body mass index on clinical outcomes associated with percutaneous nephrolithotomy. Can Urol Assoc J 7(April):1–5 49. Sari E, Tepeler A, Yuruk E, Resorlu B, Akman T, Binbay M, Armagan A, Unsal A, Muslumanoglu AY (2013) Effect of the body mass index on outcomes of flexible ureterorenoscopy. Urolithiasis 41(6):499–504 50. Thomas R, Cass S (1993) Extracorporeal shock wave lithotripsy in morbidly obese patients. J Urol 150(1):30–32 51. Hippisley-Cox J, Coupland C (2010) Predicting the risk of chronic kidney disease in men and women in England and Wales: prospective derivation and external validation of the QKidney scores. BMC Fam Pract 11:49 52. Alexander RT, Hemmelgarn BR, Samuel S, Klarenbach SW, Curhan GC (2012) Kidney stones and kidney function loss: a cohort study. Open Access 5287:1–8 53. Bellman GC, Davidoff R, Candela J, Gerspach J, Kurtz S, Stout L (1997) Tubeless percutaneous renal surgery. J Urol 157(5):1578–1582 54. Istanbulluoglu MO, Ozturk B, Gonen M, Cicek T, Ozkardes H (2009) Effectiveness of totally tubeless percutaneous nephrolithotomy in selected patients: a prospective randomized study. Int Urol Nephrol 41(3):541–545 55. Sivalingam S, Cannon ST, Nakada SY (2013) Current practices in percutaneous nephrolithotomy among endourologists. J Endourol 56. Turk C, Knoll T, Petrik A, Sarica K, Skolaricos A, Straub M, Seitz C (2001) Guidelines on urolithiasis. Eur Urol 40(4):362–371 57. Tyson MD, Humphreys MR (2014) Postoperative complications after percutaneous nephrolithotomy: a contemporary analysis by insurance status in the United States. J Endourol 28(3):291–297 58. Michel MS, Trojan L, Rassweiler JJ (2007) Complications in percutaneous nephrolithotomy. Eur Urol 51(4):899–906 59. Birkmeyer JD, Stukel TA, Siewers AE, Goodney PP, Wennberg DE, Lucas FL (2003) Surgeon volume and operative mortality in the United States. N Engl J Med 349(22):2117–2127

World J Urol 60. Konety BR, Dhawan V, Allareddy V, Joslyn S (2005) Impact of hospital and surgeon volume on in-hospital mortality from radical cystectomy: data from the health care utilization project. J Urol 173(5):1695–1700 61. Kadlec AO, Ellimoottil C, Guo R, Trinh Q-D, Sun M, Turk TM (2013) Contemporary volume–outcome relationships for percutaneous nephrolithotomy: results from the nationwide inpatient sample. J Endourol 27(9):1107–1113

62. Thomas K, Smith NC, Hegarty N, Glass JM (2011) The Guy’s stone score-grading the complexity of percutaneous nephrolithotomy procedures. Urology 78(2):277–281 63. Opondo D, Gravas S, Joyce A, Pearle M, Matsuda T, Sun YH, Assimos D, Denstedt J, de la Rosette J (2014) Standardization of patient outcomes reporting in percutaneous nephrolithotomy. J Endourol 28(7):767–774

13