World Journal of Urology

, Volume 28, Issue 4, pp 531–535

External validation of the preoperative anatomical classification for prediction of complications related to nephron-sparing surgery


    • Department of UrologyMedical University of Vienna
  • Sandra Waalkes
    • Department of UrologyMedical School of Hannover
  • Tobias Klatte
    • Department of UrologyMedical University of Vienna
  • Markus A. Kuczyk
    • Department of UrologyMedical School of Hannover
  • Peter Weibl
    • Department of UrologyMedical University of Vienna
  • Gerd Schüller
    • Department of RadiologyMedical University of Vienna
  • Axel S. Merseburger
    • Department of UrologyMedical School of Hannover
  • Mesut Remzi
    • Department of UrologyMedical University of Vienna
Original Article

DOI: 10.1007/s00345-010-0577-8

Cite this article as:
Waldert, M., Waalkes, S., Klatte, T. et al. World J Urol (2010) 28: 531. doi:10.1007/s00345-010-0577-8



Ficarra et al. (Eur Urol 56:786–793, 2009) published a preoperative anatomical classification (PADUA) to assess the impact of anatomical parameters of renal tumors on complication rate of nephron-sparing surgery (NSS). The objective of this study is to provide a bi-center external validation of this classification using the technique of hilar arterial clamping during open and laparoscopic NSS and to correlate the PADUA score to the ischemia time.


240 consecutive tumors treated with open and laparoscopic NSS were reclassified according to the PADUA classification. Complications were graded according to the modified Clavien system (Dindo et al. in Ann Surg 240:205–213, 2004). Chi-square tests and multivariate logistic regression models addressed the predictive value of PADUA classification on overall complication rate and grade.


Mean patient age was 62.2 ± 13.3 years. Eastern Cooperative Oncology Group performance was 0 in 76%, 1 in 22% and 2 in 2%. 61 (25%) were treated laparoscopically. The median PADUA score was 7.5 (range 6–13). Mean surgery and ischemia time was 189 ± 95 and 24 ± 22 min, respectively. Overall complication rate was 23% (n = 54). On univariate analysis, the PADUA score correlated with complication rate (p < 0.001) of open and laparoscopic NSS. On multivariate, only the PADUA score correlated with complication rate (p = 0.0056). Ischemic time correlated with the PADUA score and was significantly higher in PADUA score ≥ 10 (p = 0.034).


The PADUA score is a reliable tool to preoperatively predict the risk of complications. In addition, it might be beneficial for a more objective patient selection for laparoscopic surgery and teaching NSS.


Anatomical classificationIschemic timeLearning curveNephron-sparing surgeryRenal cell cancer


For renal cell cancer ≤ 4 cm, nephron-sparing surgery (NSS) has become the standard treatment of choice, due to excellent oncological outcomes (5–10-year cancer-specific survival rates of 95–100%) and lower negative impact on long-term kidney function [36]. However, open and laparoscopic NSS is a challenging procedure with a complication rate of about 20–25% [36]. As NSS has become the standard treatment of small renal masses, it is obvious that more objective parameters to further classify a renal mass are needed. Which tumors can be safely operated with a low complication rate, which tumors are suitable for a teaching operation and which tumors can be done laparoscopically? Moreover, studies reported the feasibility of NSS in selected tumors of a size from 4.1 to 7 cm [7, 8]. So, anatomical aspects other than tumor size were considered as parameters determining treatment possibilities and two classification systems have been introduced recently [1, 9]. The R.E.N.A.L. nephrometry score characterizes tumors using size, growth pattern, location and nearness to the renal sinus or collecting system. Low- and moderate-scored tumors often underwent minimal invasive NSS whereas high-scored tumors were more likely to undergo open partial or laparoscopic radical nephrectomy [9]. The PADUA classification uses a similar approach. It rates tumors and allocates points according to the following anatomical aspects: polar location, exophytic growth pattern, and location at the renal rim, involvement of the renal sinus or urinary collecting system and tumor size. Ficarra et al. [1] concluded that the PADUA score is a simple anatomical system that can be used to predict the risk of surgical and medical perioperative complications in patients undergoing open NSS in cancers up to a size of 7 cm. However, as the original study population only consisted of patients who underwent open NSS with manual or clamp compression of the renal parenchyma, the applicability of the score could be limited.

The aim of this study is to (1) provide external validation for the PADUA classification; (2) to determine whether it is applicable to lesions treated with open or laparoscopic NSS in cold and warm ischemia; and (3) test its ability to predict operation time, ischemia time and intraoperative blood loss.

Materials and methods

We retrospectively evaluated the data of 297 consecutive patients who underwent NSS at two institutions (Medical University Vienna and Medical School Hannover) between 2006 and 2009. The study was approved by the ethics committee of the Medical University of Vienna. All patients underwent laparoscopic or open NSS for clinical T1 tumors and were staged preoperatively with computed tomography (CT) or magnetic resonance imaging (MRI). Thirty-four patients from Vienna and 23 patients from Hannover were excluded, because the CT scan was not available for independent review (n = 26) or more than one tumor was unilaterally operated in one session in case of multifocality. For the purpose of this study, all images were reclassified according to the PADUA classification by two urologic surgeons and one uro-radiologist from Vienna and two urologic surgeons from Hannover, all blinded to the final outcome of the patients. A consensus regarding the final PADUA score for each tumor was reached between all observers. The parameters of the PADUA classification were assigned to each tumor and points allocated accordingly: (1) anterior or posterior face of the kidney; (2) polar location: superior/inferior or middle; (3) tumor deepening into the parenchyma: ≥50% exophytic, ≤50% exophytic, endophytic; (4) rim location: lateral or medial; (5) renal sinus involvement: involved or not involved; (6) involvement of the urinary collecting system: not involved, dislocated/infiltrated and (7) tumor size: ≤4, 4.1–7, >7 cm [1].

All patients underwent transperitoneal laparoscopic (n = 61) or open extraperitoneal NSS (n = 179) by six experienced urologic surgeons. Arterial clamping was performed in all cases. In open NSS, cold ischemia was used whereas in laparoscopic NSS warm ischemia was applied. The tumors were removed with an adequate safety margin in all cases. Surgery time was calculated from skin incision to wound closure. Ischemic time was recorded for all cases and calculated from the clamping of the renal artery until the removal of the clamp.

Information on intra- and postoperative complications were extracted from patient records until 40 days after surgery. Standard follow-up procedures were a hospital stay up to 7 days after surgery and controls 14 and 28–30 days after surgery on an outpatient basis. All complications were graded according to the modified Clavien system published in 2004 [2].

Statistical analysis

Continuous variables are presented as mean ± standard deviation. For univariate analysis, the Student’s t test was performed. Categorical variables were compared using Chi-square tests. A multivariate logistic regression model was fit to identify independent preoperative factors associated with the risk of complication. A p < 0.05 was considered statistically significant. Data were analyzed using R v.2.8.0 Software.


Two hundred forty tumors treated with NSS were included in this study. One hundred sixty-three (68%) patients were male. Mean patient age was 62.2 ± 13.3 years. Eastern Cooperative Oncology Group (ECOG) performance status was 0 in 182 (76%), 1 in 53 (22%) and 2 in 5 (2%) patients. Mean tumor size was 3.4 ± 2.8 cm. Sixty-three (26%) were greater than 4 cm on preoperative CT scan. Histology was benign in 40 pts (17%), clear cell carcinoma in 154 pts (64%), papillary carcinoma in 22 pts (9%), chromophobe carcinoma in 20 pts (8%) and unclassifiable in 4 pts (2%). TNM stage was pT1 a/b in 147 (87%), pT2 in 1 (0.5%) and pT3a in 25 (12.5%). Complications were observed in 54 patients (22.5%). Complications stratified according to the modified Clavien system were Clavien 0 in 77.5%, 1 in 10.8%, 2 in 6.3%, 3a in 2.5%, 3b in 2.1% and 4 in 0.8% of patients, respectively. One hundred seventy-nine (75%) tumors were treated with open NSS, whereas 61 (25%) were removed laparoscopically. Mean operative time was 189 ± 95 min, mean ischemic time 24 ± 22 min and mean blood loss 330 ± 240 ml. Overall complications, mean operative time, ischemic time and blood loss were not statistically significant different in the open or laparoscopic group (p = 0.699, 0.634 and 0.480, respectively). Only 13% (8 pt) had an ischemic time > 30 min in the laparoscopic NSS group.

The anatomical features stratified according to the PADUA score are given in Table 1. The median value of PADUA score was 7.5 (range 6–13). Tumor size and PADUA scores were significantly smaller and lower in the laparoscopic group (median 8 vs. 6, p = 0.008, Chi-square test). It was able to predict the overall risk of complications for all patients (p < 0.001) but did not correlate with the Clavien score (p = 0.567), thus the PADUA score was not able to predict the severity of complications. In the subgroup with a tumor size greater 4 cm, the PADUA score was also a significant predictor of overall complication rate (p = 0.021). The PADUA score was able to predict complications in the open as well as in the laparoscopic group (p = 0.028 and 0.031). A higher PADUA score significantly predicted a higher overall operating time and ischemic time (p = 0.024 and 0.034, respectively), the significance for ischemic time being most prominent in tumors with a PADUA score ≥ 10 (22 ± 12 vs. 34 ± 42 min; p = 0.006) but failed to predict a higher blood loss (p > 0.05).
Table 1

General anatomical features of the tumors according to the PADUA classification [1] and their predictiveness for overall complications


Number of patients (%)

Patients without complications/Clavien 0 (n = 186)

Patients with complications/Clavien ≥ 1 (n = 54)

Univariate analysis, p value

Age (years), mean ± SD


62 ± 12

63 ± 14


Blood loss (ml), mean ± SD


307 ± 230

392 ± 264


Open NSS (no. of patients, %)


136 (76%)

50 (82%)


Laparoscopic NSS (no. of patients, %)


43 (24%)





139 (57.9%)





101 (42.1%)




Polar location




140 (58.3%)





100 (41.7%)




Exophytic rate



 ≥50% of tumor

157 (65.4%)




 <50% of tumor

68 (28.3%)




 Endophytic tumor

15 (6.3%)




Renal rim



 Lateral rim

185 (77.1%)




 Medial rim

55 (22.9%)




Renal sinus



 Not involved

191 (79.6%)





49 (20.4%)




Urinary collecting system



 Not involved

168 (70%)





72 (30%)




Tumor size (cm)




177 (73.8%)





55 (22.9%)





8 (3.3%)




NSS nephron sparing surgery

Considering correlation between overall complications rate and the single factors investigated in this study, only involvement of the renal sinus and collecting system and tumor size stratified in ≤4, 4.1–7 and >7 cm showed statistical significance (p = 0.009, 0.001 and 0.016, respectively; Table 1). On multivariate analysis, PADUA score was an independent predictor of the occurrence of complications (OR 1.34; 95% CI 1.14–1.59; p = 0.001) especially when comparing patients with a PADUA score < 10 and ≥10 (OR 3.08; 95% CI 1.55–6.11; p = 0.001; Table 2).
Table 2

Multivariate analysis of factors predicting complications after nephron-sparing surgery (NSS)


Risk ratio

95% confidence interval

p value





Open vs. laparoscopic NSS




PADUA score





Until recently studies relating to NSS of solid enhancing renal tumors only contained sparse information regarding relevant anatomical features of these lesions [1013]. As the details provided were often incomplete, the results of the different studies are not comparable and no general conclusions can be made. The main anatomical features reported usually were tumor size and endophytic growth pattern, but objective comparable data were lacking. Recently, the PADUA classification was introduced to provide an easy way to compare renal lesions ≤ 7 cm in terms of anatomical features. It also offers a way to predict the risk of complications after and related to NSS and aims to improve and individualize selection criteria for the available surgical approaches, i.e. open or laparoscopic NSS or minimal invasive ablative techniques. Tumors are scored with 6–13 points taking into consideration the maximal diameter of the tumor and five anatomical aspects: the polar location, exophytic growth rate, location at the renal rim and involvement of the renal sinus and urinary collecting system.

In the original study, 164 patients were included. All tumors were removed in open NSS without clamping of the main vessels but with local manual compression or parenchymal clamping [1]. In the original group, the PADUA score was able to predict the risk of overall complications (p < 0.001) more accurately than clinical tumor size in uni- and multivariate analysis. Limitations of the original study were foremost the lack of evaluation of warm ischemia time. All patients were treated by open partial nephrectomy. Studies have shown that the oncological results of laparoscopic NSS equal those of open NSS and the method is widely used for selected renal tumors [10, 14]. So, it is of potential interest if the PADUA score is also applicable for laparoscopic NSS and if it could be used as a tool to select patients for this procedure.

In our study, we provide an external bi-center validation of the PADUA scoring system. Two hundred forty patients were included. In terms of age, gender, TNM stage and the PADUA sub-classification, our series was comparable to the study of Ficarra et al [1]. The main differences were in the used surgical techniques. In contrast to the original paper, 75% of our patients were operated with open NSS with clamping of the renal artery. The remaining 25% of tumors were removed laparoscopically. We were able to show that the PADUA score is also applicable in patients treated with laparoscopic NSS and accurately predicts overall complications in this group. We also showed that the PADUA score is usable in open NSS with clamping of the renal artery as means of perfusion control. Moreover, we found that the PADUA score was significantly lower in the laparoscopic group implying its further use as a selection tool for (1) laparoscopic NSS, (2) teaching purposes, and (3) for upcoming studies concerning minimal invasive procedures, thereby ruling out selection biases that may influence the results. This is even more important since recent studies suggested an influence of local tumor anatomy on pathology and subsequently on prognosis. Venkatesh et al. [13] reported that in most exophytic tumors 45% were benign and only 4% high grade.

The PADUA score failed to predict the severity and type of complications. This could be due to the limited patient and complication number in our study. To investigate its usefulness in predicting these factors, studies with larger patient populations are needed.

Another hypothesis, which Ficarra et al. could not validate, was whether a higher PADUA score predicts a higher ischemia time. In our study, population data on overall surgery time, ischemia time and intraoperative blood loss were available. All three variables were not influenced by surgical approach. The PADUA score predicted a higher ischemia time in both the laparoscopic and open group. It also predicted a higher overall operation time. Tumors with a score of 6–7 were operated on significantly shorter than tumors ≥ 8. Not surprisingly, the PADUA score could not predict blood loss, because this parameter is only minimally influenced by the anatomical characteristics of the tumor in NSS, because of clamping of the renal artery. We observed that most bleeding complications are due to problems during hilar preparation, and general blood loss during tumor enucleation itself is low. However, postoperative bleeding mainly due to arterio-venous fistulas was correlated to anatomical features. In Vienna, two cases had a postoperative bleeding complication and super selective embolisation was necessary.

We also provide further insights regarding the validity of the PADUA score in T1b tumors potentially suitable for NSS. In our cohort, 63 patients (26%) harbored tumors > 4 cm, nearly three times as much as in the original paper. Seven (11%) of them were treated by laparoscopic NSS. With increasing size, we also observed a rise in complications (22% in tumors ≤ 4 cm vs. 32% in tumors 4.1–7 cm—however mainly Clavien I). This fact is also reflected that in contrast to Ficarra et al.’s study tumor size alone was a predictor of overall complications in univariate analysis. But even in tumors greater than 4 cm in diameter, the PADUA score was a reliable predictor of complications and could function as a tool for selecting tumors > 4 cm suitable for NSS.

A drawback of our study is the retrospective character. In addition, no patients underwent robotic NSS.


The PADUA score is a reliable tool to preoperatively predict the risk of complications and important parameters such as ischemic time. It is applicable to renal masses treated with open and laparoscopic NSS. In addition, it can help clinicians in selecting patients suitable for laparoscopic surgery or teaching NSS.

Conflict of interest statement

The authors declare that they have no conflict of interest.

Copyright information

© Springer-Verlag 2010