Original Paper
Urologia Urol Int
Internationalis DOI: 10.1159/000504145
Selecting the Best Elements from Previous Kidney
Tumor Scoring Systems to Restructure Efficient
Predictive Models for Surgery Type
Huijiang Zhang Zhaoyu Xu Xuedong Chen Yongchun Li Peng Li
Weili Zhang Junjie Ye
Department of Urology, People’s Hospital of Lishui, Lishui, China
Keywords
Kidney tumor · Scoring systems · Surgery type
Abstract
Objective: The aim of this work was to select the best ele-
ments from previous scoring systems to restructure efficient
predictive models for surgery type. Methods: Sixteen ele-
ments were selected from 7 systems (RENAL, PADUA, DAP,
ZS, NephRO, ABC, and CI). They were divided into 6 catego-
ries (tumor max. size, exophytic/endophytic, correlation
with collecting system or sinus, tumor location, contact situ-
ation with the parenchyma, invasion depth). Three elements,
selected from 3 different categories, were integrated to es-
tablish a total of 320 new models. According to AUC rank,
optimized models were developed, and these models were
divided into 3 sections. An analysis of the distribution of
the 6 categories was made to explore the predictive capaci-
ties of the models. Results: A total of 166 consecutive pa-
tients were included. Seventy-five patients underwent radi-
cal nephrectomy operations. The AUC of the 7 systems
ranged from 0.81 to 0.844. Three optimized models (AUC
0.88) were developed to predict surgery type. These opti-
mized models were composed of DAP (D), PADUA, (sinus),
and ABC; DAP (D), RENAL (N), and ABC; NePhRO (O), PADUA
(UCS), and ABC. Two categories (“exophytic/endophytic,” p <
0.001; “correlation with collecting system or sinus,” p = 0.001)
© 2019 S. Karger AG, Basel
E-Mail karger@karger.com
www.karger.com/uin
Received: April 24, 2019
Accepted after revision: October 14, 2019
Published online: November 20, 2019
were nonuniformly distributed. Conclusions: Seven systems
held good predictive power for surgery type. Three opti-
mized models were developed. “Correlation with collecting
system or sinus” is a critical factor for predicting surgery type.
© 2019 S. Karger AG, Basel
Introduction
With the growing use of ultrasound and computed to-
mography (CT), an increasing number of kidney tumors
have been diagnosed accidentally. Surgery has become
one of the major treatments for these tumors. A challenge
for the urologic community is to minimize lesions and
maximize the long-term renal function of the patients;
multiple factors have to be taken into account. However,
the most important factor is the features of the tumor.
Tumor scoring systems have provided an objective and
accurate method to discuss tumor anatomical com-
plexity.
The RENAL (radius, exophytic/endophytic, nearness to
collecting system or sinus, anterior/posterior and location
relative to polar lines) nephrometry score [1] was the first
system to semi-quantitate important anatomical dimen-
sions of the renal mass and was developed in 2009. Over the
H.Z., Z.X., X.C., and Y.L. are co-first authors.
130.209.6.61 - 11/30/2019 10:20:56 AM
Junjie Ye
Department of Urology
People’s Hospital of Lishui
Dazhong Street No. 14, Lishui, Zhejiang 323000 (China)
Glasgow Univ.Lib.
Downloaded by:
E-Mail 905940344 @ qq.com
last 10 years, 10 such systems have been reported, such as
the PADUA (preoperative aspects and dimensions used
for an anatomical) [2], CI (centrality index) [3], DAP (di-
ameter-axial-polar nephrometry) [4], ABC (arterial-based
complexity) [5], NePhRO (nearness physical, zones, radius,
organization) [6], and MAP (Mayo adhesive probability
score) [7]. Although some predictive abilities of these sys-
tems have been validated [8], none of them have been au-
thoritatively recommended. Some experts have comment-
ed that a few studies have evaluated the actual value of the
single components included in the scores [9]. Additionally,
reducing the amount of small or insignificant parameters
could enhance predictive power [10]. Therefore, we want to
select the most useful components from the existing sys-
tems to develop a more concise and efficient model.
Patients and Methods
Cohort
Consecutive patients with a renal tumor and aged older than 18
years were included in this study. Patients who met one of the fol-
lowing conditions were excluded: (1) metastatic carcinoma; (2)
multiple tumors in one kidney; (3) an organic or functionally soli-
tary kidney. All surgeries and other treatments occurred at the
same institution.
Methods
Seven scoring systems (RENAL, PADUA, DAP, ZS [11],
NePhRO, ABC, and CI) were gathered for analysis. This study can
be divided into 3 parts. First, we compared the predictive powers of
these 7 systems based on the receiver operating characteristic
(ROC) curve. Second, 3 tumor elements were used to rebuild opti-
mized models to predict surgery type. To obtain these elements, all
the components of the abovementioned systems were analyzed.
This study includes the numeric components of the CI (X and Y)
instead of the CI itself because it is a processed character, calcu-
lated from 2 primal factors. Identical components between differ-
ent scoring systems were taken as one element. In all, 22 numeric
or descriptive elements were collected for further analysis. Through
univariable analysis, 16 favorable elements were selected. Accord-
ing to their definitions, they were divided into 6 categories. Three
elements selected from 3 different categories were grouped and in-
tegrated to establish a new model. This resulted in 320 groups, and
their predictive power was assessed to elect the optimized systems.
Third, we assessed each category’s capability for accurate predic-
tion. The 320 groups were divided into trisections depending on the
AUC rank. An analysis of the distribution of these 6 categories in
the 3 sections was performed to explore their predictive capacities.
Each system or component was assessed based on CT (Aquil-
ion ONE TSX-301A, a 320-detector row CT scanner) or MRI (Dis-
covery MR750 3.0T) images. In the study, CT and MRI materials
were regarded as equivalent, and all the images were contrast en-
hanced. A single radiologist, avoiding the discordancy, reviewed
all the cases blindly and independently. The images were analyzed
according the accurate definitions of each element. Priority was
given to the contrast-enhanced image scan to obtain dates. If there
2 Urol Int
DOI: 10.1159/000504145
were some doubts, the reader would scan the images repeatedly
and combine with the non-contrast-enhanced image to choose the
most appropriated measurement point.
Statistical Analysis
ANOVA was used to select possible predictors of the surgery
type from 22 elements, and χ2 was used to analyze the distribution
of the 6 categories. p < 0.05 was regarded as a significant difference.
ROC curves were used to compare the predictive abilities of each
system and elements. All statistical analyses were performed using
SPSS version 23.
Results
A total of 166 consecutive patients (78 males; 88 fe-
males; mean age 55.6 years; mean tumor size 47.8 mm)
were collected. Of these, 75 (45.2%) patients underwent
radical nephrectomy operations, and the other 91 (54.8%)
underwent partial nephrectomy operations. The RENAL,
PADUA, DAP, ZS, NephRO, ABC, and CI scoring sys-
tems were included in this study. The AUC of these 7 sys-
tems ranged from 0.81 to 0.844 (Fig. 1), with the ZS scor-
ing system performing the best.
We extracted 22 elements from the above 7 systems.
According to the univariate analysis (Table 1), the corre-
lation between the 16 elements and surgery type were sta-
tistically significant. The 16 elements were divided into 6
categories according to their definitions: tumor max. size
(3 elements), exophytic/endophytic (1 element), correla-
tion with collecting system or sinus (3 elements), tumor
location (5 elements), contact situation with the paren-
chyma (1 element), and invasion depth (3 elements; Table
2). Three elements were selected from 3 different catego-
ries to generate a new model. The definitions and scoring
method of all elements were not changed. Overall, 320
new models were developed. The AUC of these new mod-
els ranged from 0.7 to 0.88 (Fig. 2).
Three optimized models (AUC 0.88) showed the most
power in predicting surgery type (Fig.  1). Optimized
model 1 was composed of DAP (D), PADUA (sinus), and
ABC; optimized model 2 was composed of DAP (D), RE-
NAL (N), and ABC; and optimized model 3 was com-
posed of NephRO (O), PADUA (UCS), and ABC; in ad-
dition, there were 3 special kinds of models that were re-
structured with some elements of RENAL, PADUA, or
NephRO. We called these special models as restructured
RENAL (AUC 0.847; Fig. 3), restructured PADUA (re-
structured PADUA1, AUC 0.828; restructured PADUA2,
AUC 0.806; Fig. 4), and restructured NephRO (restruc-
tured NephRO1, AUC 0.802; restructured NephRO2,
130.209.6.61 - 11/30/2019 10:20:56 AM
Zhang/Xu/Chen/Li/Li/Zhang/Ye
Glasgow Univ.Lib.
Downloaded by:
 Color version available online
60

40

Frequency
20

0
0.65 0.70 0.75 0.80 0.85 0.90
AUC value
Fig. 1. The distribution of the AUC values
of the 320 restructured models.

Table 1. Univariate analysis of 22 elements from the 7 systems

RENAL (R)/ RENAL (E)/ RENAL (N) RENAL (A) RENAL (L) PADUA (ap) PADUA PADUA PADUA PADUA (UCS)
PADUA (max. PADUA (longitudinal (rim) (sinus)
diameter) (endophytic location)
properties)

F 20.806 2.095 72.793 1.431 30.798 0.027 13.476 2.023 99.668 97.101
p value <0.0001 0.150 <0.0001 0.233 <0.0001 0.869 <0.0001 0.157 <0.0001 <0.0001
DAP (D) DAP (A) DAP (P) ZS (Ri) ZS-physical NePhRO (Ne)/ NePhRO (Ph) NePhRO (R) NePhRO (O) ABC CI (X) CI (Y)
location ZS (depth)

F 41.052 25.633 28.276 42.861 63.463 76.283 34.315 31.355 3.869 78.824
p value <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 0.05 <0.0001 0.076 0.073

Table 2. Sixteen elements divided into 6 categories according to their definitions

Category Tumor max. size Exophytic/ Correlation Tumor location Contact Invasion depth
endophytic with UCS or sinus situation with
the parenchyma

Component RENAL (R)/PADUA NePhRO (O) RENAL (N); RENAL (L); ZS (Ri) ABC;
(max. diameter); PADUA (sinus); PADUA (longitudinal DAP(A);
DAP (D)/NePhRO (R) PADUA (UCS) location); NePhRO (Ne)/
DAP (P); NephRO (Ph)/ ZS (depth)
ZS (physical location)
130.209.6.61 - 11/30/2019 10:20:56 AM

Predictive Kidney Models for Surgery Urol Int 3

Type DOI: 10.1159/000504145
Glasgow Univ.Lib.
Downloaded by:
 Color version available online

Color version available online

ROC curve ROC curve
1.0 1.0

Source of the curve

RENAL
0.8 Source of the curve 0.8 Restructured RENAL
RENAL
PADUA
DAP
ZS
0.6 NePhRo 0.6
ABC

Sensitivity
Sensitivity

CI
Optimized model 1
Optimized model 2
Optimized model 3
0.4 0.4

0.2 0.2

0 0
0 0.2 0.4 0.6 0.8 1.0 0 0.2 0.4 0.6 0.8 1.0
1 – specificity 1 – specificity

Fig. 2. The ROC curves of the RENAL, PADUA, DAP, ZS, Fig. 3. The ROC curves of RENAL and restructured RENAL; the
NePhRO, ABC, and CI models and the 3 optimized models: the AUCs of the 2 systems were 0.83 and 0.847, respectively.
AUCs of these systems were 0.83, 0.835, 0.825, 0.844, 0.842, 0.81,
0.831, 0.88, 0.88, and 0.878, respectively.

AUC 0.828; restructured NephRO3, AUC 0.847; restruc-

Color version available online

ROC curve
1.0 tured NephRO4, AUC 0.847; Fig. 5).
According to the AUC rank, these 320 restructured
Source of the curve models were divided into 3 equal levels. The mean AUC
0.8 PADUA
Restrutured PADUA1
values of the 3 levels were 0.79 (first level), 0.829 (second
Restrutured PADUA2 level), and 0.851 (third level), which were significantly
different (F = 441, p < 0.001). The distribution of the ele-
0.6 ments was analyzed. Four categories (tumor max. size,
Sensitivity

tumor location, contact situation with the parenchyma,

invasion depth) were equally distributed. Two categories
0.4 (“exophytic/endophytic,” χ2 = 46.8, p < 0.001; “the cor-
relation with collecting system or sinus,” χ2 = 14.32, p =
0.001) had no uniform distribution (Table 3). The first
0.2 level contained 45.2% of the category “the correlation
with collecting system or sinus,” and the third level con-
tained 67.4% of the category “exophytic/endophytic.”
0
0 0.2 0.4 0.6 0.8 1.0
1 – specificity Discussion

Fig. 4. The ROC curves of PADUA, restructured PADUA1, and A series of similar renal tumor models have been intro-
restructured PADUA2; the AUCs of the 3 systems were 0.837, duced over the last decade. These models have revealed a
0.828, and 0.806, respectively. close association with ischemia time, operation time, com-
130.209.6.61 - 11/30/2019 10:20:56 AM

4 Urol Int Zhang/Xu/Chen/Li/Li/Zhang/Ye

DOI: 10.1159/000504145
Glasgow Univ.Lib.
Downloaded by:
 Table 3. Analysis of the distribution of the 16 elements

Tumor Exophytic/ Correlation with Tumor The contact situation Invasion

max. size endophytic UCS or sinus location with parenchyma depth

First level 56 9 84 71 29 69
Second level 65 19 60 75 37 62
Third level 65 58 42 84 20 55
χ2 0.87 46.8 14.32 1.16 5.04 1.58
p value 0.668 <0.001 0.001 0.571 0.083 0.473

plexity of the surgery, perioperative complications, and re-

Color version available online

ROC curve
nal lesions in the literature [8]. Moreover, some of these 1.0
models show a statistically significant correlation with
pathological type [12, 13] and tumor growth rates in pa-
tients under active surveillance [14]. A number of studies 0.8
have illustrated the superiority of these models in different Source of the curve

circumstances. MAP (Mayo adhesive probability score) NePhRO

Restrutured NePhRO1
was shown to explain the correlation between potential 0.6
Restrutured NePhRO2
Restrutured NePhRO3
clinical outcomes and adherent perinephric fat [15]. The Sensitivity
Restrutured NePhRO4

ABC scoring system was found to predict the clinical re-

sults of partial nephrectomy based on the relationship be-
tween the arterial vascular anatomy and the tumor [5].
DAP nephrometry is an integration and optimization of
RENAL and the centrality index scoring systems [4]. On
the other hand, there were some negative research results.
Maxwell AWP found that the maximum tumor diameter
demonstrated better performance than existing tumor
scoring systems for predicting local tumor recurrence after
renal cell carcinoma ablation [16]. Another study showed
that the reproducibility between the radiologist and the
urologist was not perfect, especially for some items relating
to the tumor location [17]. Gu et al. [18] considered that
the ABC scoring system cannot predict the occurrence of
complications and postoperative renal function.
Considering the patients, how much benefit they can
get from surgical treatment is first decided by the surgery
type, and so the surgeons need an accurate prediction to
choose the most suitable treatment. Therefore, we think
the most important task for scoring systems is to produce
a better prediction for surgery type. For surgeons, a sim-
ple tool would be convenient to use in their daily work.
Therefore, simplifying the scoring systems should be giv-
en more attention. We determined that 1 or 2 elements
would be too simple to describe the feature of the tumor
anatomy, and 4 elements would be difficult to manage to
build the model. Therefore, in this study, we attempted to
choose the 3 best elements from the previous systems to
restructure an optimized model.
0.4
0.2
0
0 0.2 0.4 0.6 0.8 1.0
1 – specificity
Fig. 5. The ROC curves of NePhRO, restructured NePhRO1, re-
structured NePhRO2 restructured NePhRO3, and restructured
NePhRO4; the AUCs of the 5 systems were 0.843, 0.802, 0.828,
0.847, and 0.847, respectively.
A number of researchers have performed similar stud-
ies. Tobert et al. [10] chose and combined 4 variables
from the RENAL, PADUA, and CI nephrometry systems
to rebuild a new scoring system which gains a better pre-
dictive ability for surgery type. However, CI is not an
original variable but is calculated from 2 anatomic fac-
tors. This increased the calculation difficultly for the new
system. Roscigno et al. [19] found that the predictive ac-
curacy of nephrometric scores could be improved by add-
ing clinical patient characteristics. He used PADUA, RE-
NAL, CCI, and BMI to generate an algorithm (the Rosco
score) to efficiently predict complications. However, the
130.209.6.61 - 11/30/2019 10:20:56 AM

Predictive Kidney Models for Surgery Urol Int 5

Type DOI: 10.1159/000504145
Glasgow Univ.Lib.
Downloaded by:
Rosco score involved some repetitive elements because
PADUA and RENAL contain a number of similar ana-
tomic elements, and the calculation was also complex.
In our study, we included most of the systems and tumor
features from the literature. Some systems, such as MAP,
which focus on perinephric conditions, were excluded.
Based on professional experience, these systems had little
correlation with surgery type. Overall, each of the 7 systems
included hold good predictive ability for surgery type. The
predictive power of ABC (AUC 0.81) was slightly weaker
than that of the others (AUC 0.83–0.84). The predictive
power of RENAL (AUC 0.83), PADUA (AUC 0.837), and
CI (AUC 0.834) was lower than that in similar studies (RE-
NAL, AUC 0.9; PADUA, AUC 0.88; CI, AUC 0.91) [10].
This may be because surgeons preferred radical nephrec-
tomy for treating complex tumors in our institution.
We succeeded in producing 3 optimized models that
were simpler to use and possessed a stronger predictive
power than that of previous systems. We are confident
that these 3 simple systems will be appreciated by sur-
geons and will benefit from widespread use. Additionally,
there were 3 special kinds of models that were restruc-
tured by parts of elements of the previous scoring sys-
tems. We were surprised to find that the predictive pow-
er of these kinds of models was equivalent to, and some-
times even exceeded, that of the previous models: for
example, RENAL (AUC 0.83) versus restructured RE-
NAL (AUC 0.847), and NephRO (AUC 0.843) versus re-
structured NephRO 3 (AUC 0.847). This finding indicats
that previous scoring systems include too many elements,
some of which are not helpful in predicting surgery type.
Not all of the tumor characteristics provided equal pre-
dictive power for surgery type. In this study, most of the
“exophytic/endophytic” category was in the third level of
the AUC rank, and the majority of “the correlation with
collecting system or sinus” category was in the first level.
It can be speculated that the correlation with the collecting
system or sinus was more important for predicting sur-
gery type, while the growth pattern of “exophytic/endo-
phytic” was least important. Hence, surgeons should pay
more attention to “the correlation with collecting system
or sinus” elements even though its total score is not high.
An ideal system should meet 2 requirements: an accu-
rate predictive power and ease of use and comprehension.
Urologic experts have sought to integrate the best-per-
forming items into a model and minimize the number of
items as best as possible, despite the difficulty of this task.
However, these 2 requirements are self-contradictory.
Recent research has indicated that including more pa-
rameters may influence subjective appraisal [20]. The C-
6 Urol Int
DOI: 10.1159/000504145
index and DAP score were calculated relatively simply, so
they are more reproducible compared with other systems
[20]. Another study demonstrated that adding clinical
patient characteristics can improve the predictive accu-
racy of nephrometric scores [19]. Therefore, we boldly
hypothesize that we can develop 2 versions of the model.
Version 1 would possesses an outstanding power to de-
scribe tumor features and predict clinical outcomes and
would be applied by the urologic community frequently
for renal tumor studies and academic exchange. Version
2, which would be practical and intuitive, could be easily
used in daily work, although it would not be as robust as
version 1 in its predictive capability.
A nephrometry scoring system is a useful tool for tumor
descriptions, treatments, and the prediction of prognosis.
Hitherto, different counties, hospitals, and research teams
may use different models due to the lack of an authoritative
recommendation. To improve the present situation, first,
a large-scale survey is needed to give an overview of actual
usages for each model and make inquiries about whether
these scoring systems are useful in surgeons’ daily work
and study. Second, surgeons groups can integrate the above
information to make an appropriate guide for urologic
groups. To the best of our knowledge, there is no relevant
study that has attempted to develop guidelines for recom-
mended scoring system use. The EU guideline only indi-
cates that PADUA, RENAL, and CI have been proposed,
but without a recommendation for usage [21].
Our study includes the following limitations: (1) it is a
retrospective and single-institution study, and thus has a
selection bias; (2) a limited number of cases were includ-
ed in this study, and (3) we did not perform a larger,
multi-institutional analysis to verify the optimized mod-
els. Furthermore, we regret that we did not distinguish
CT materials from the MRI materials. Nevertheless, we
regard this study as a meaningful work, as it provides 3
more simplified and more efficient models and confirms
the importance of “the correlation with collecting system
or sinus” category to predict surgery type.
Conclusion
Seven systems (RENAL, PADUA, DAP, ZS, NephRO,
ABC, and CI) hold good predictive power for surgery
type. Three simpler and more efficient models were de-
veloped with 3 best-performing elements chosen from
the 7 systems above. Additionally “the correlation with
collecting system or sinus” is an important factor for pre-
dicting surgery type.
130.209.6.61 - 11/30/2019 10:20:56 AM
Zhang/Xu/Chen/Li/Li/Zhang/Ye
Glasgow Univ.Lib.
Downloaded by:
 Acknowledgements est of any nature in any product, service, or company that could be
construed as influencing the position presented in, or the review
Thanks to my wife, Tongtong zhou, for help with English writing. of, the manuscript.

Statement of Ethics Funding Sources

The subjects gave their informed consent and the study proto- No funding was obtained for this work.
col was approved by the institute’s committee on human research.

Disclosure Statement
We declare that we have no financial or personal relationships
with other people or organizations that could inappropriately in-
fluence our work; there is no professional or other personal inter-
Author Contributions
H.Z. developed the project and wrote the manuscript. Z.X.
wrote the manuscript. X.C., P.L., and W.Z. collected the data. Y.L.
analyzed the data. J.Y. developed the project.

References
  1 Kutikov A, Uzzo RG. The R.E.N.A.L. neph-
rometry score: a comprehensive standardized
system for quantitating renal tumor size, loca-
tion and depth. J Urol. 2009 Sep;182(3): 844–
53.
 2 Ficarra V, Novara G, Secco S, Macchi V,
Porzionato A, De Caro R, et al. Preoperative
aspects and dimensions used for an anatomi-
cal (PADUA) classification of renal tumours
in patients who are candidates for nephron-
sparing surgery. Eur Urol. 2009 Nov; 56(5):
786–93.
 3 Simmons MN, Ching CB, Samplaski MK,
Park CH, Gill IS. Kidney tumor location mea-
surement using the C index method. J Urol.
2010 May;183(5):1708–13.
  4 Simmons MN, Hillyer SP, Lee BH, Fergany
AF, Kaouk J, Campbell SC. Diameter-axial-
polar nephrometry: integration and optimi-
zation of R.E.N.A.L. and centrality index
scoring systems. J Urol. 2012 Aug; 188(2):
384–90.
  5 Spaliviero M, Poon BY, Karlo CA, Gugliel-
metti GB, Di Paolo PL, Beluco Corradi R, et
al. An Arterial Based Complexity (ABC) Scor-
ing System to Assess the Morbidity Profile of
Partial Nephrectomy. Eur Urol. 2016 Jan;
69(1):72–9.
  6 Hakky TS, Baumgarten AS, Allen B, Lin HY,
Ercole CE, Sexton WJ, et al. Zonal NePhRO
scoring system: a superior renal tumor com-
plexity classification model. Clin Genitourin
Cancer. 2014 Feb;12(1):e13–8.
  7 Davidiuk AJ, Parker AS, Thomas CS, Leibo­
vich BC, Castle EP, Heckman MG, et al. Mayo
adhesive probability score: an accurate im-
age-based scoring system to predict adherent
perinephric fat in partial nephrectomy. Eur
Urol. 2014 Dec;66(6):1165–71.
  8 Gupta R, Tori M, Babitz SK, Tobert CM, Ane-
ma JG, Noyes SL, et al. Comparison of RE-
NAL, PADUA, CSA, and PAVP Nephrome-
try Scores in Predicting Functional Outcomes
After Partial Nephrectomy. Urology. 2019
Feb;124:160–7.
  9 Volpe A, Terrone C. Anatomic classification
systems of renal tumors: new, useful tools in
renal surgical oncology. Eur Urol. 2011 Oct;
60(4):731–3.
10 Tobert CM, Shoemaker A, Kahnoski RJ, Lane
BR. Critical appraisal of first-generation renal
tumor complexity scoring systems: creation
of a second-generation model of tumor com-
plexity. Urol Oncol. 2015 Apr;33(4):167.e1–6.
11 Zhou L, Guo J, Wang H, Wang G. The Zhong-
shan score: a novel and simple anatomic clas-
sification system to predict perioperative out-
comes of nephron-sparing surgery. Medicine.
2015 Feb;94(5):e506.
12 Tay MH, Thamboo TP, Wu FM, Zhaojin C,
Choo TB, Ramaan L, et al. High R.E.N.A.L.
Nephrometry scores are associated with
pathologic upstaging of clinical T1 renal-cell
carcinomas in radical nephrectomy speci-
mens: implications for nephron-sparing sur-
gery. J Endourol. 2014 Sep;28(9):1138–42.
13 Chen SH, Wu YP, Li XD, Lin T, Guo QY,
Chen YH, et al. R.E.N.A.L. Nephrometry
Score: A Preoperative Risk Factor Predicting
the Fuhrman Grade of Clear-Cell Renal Car-
cinoma. J Cancer. 2017 Oct;8(18):3725–32.
14 Matsumoto R, Abe T, Shinohara N, Murai S,
Maruyama S, Tsuchiya K, et al. RENAL neph-
rometry score is a predictive factor for the an-
nual growth rate of renal mass. Int J Urol.
2014 Jun;21(6):549–52.
15 Bylund JR, Qiong H, Crispen PL, Venkatesh
R, Strup SE. Association of clinical and radio-
graphic features with perinephric "sticky" fat.
J Endourol. 2013 Mar;27(3):370-3.
16 Maxwell AW, Baird GL, Iannuccilli JD,
Mayo-Smith WW, Dupuy DE. Renal Cell
Carcinoma: Comparison of RENAL Neph-
rometry and PADUA Scores with Maximum
Tumor Diameter for Prediction of Local Re-
currence after Thermal Ablation. Radiology.
2017 May;283(2):590–7.
17 Benadiba S, Verin AL, Pignot G, Bessede T,
Drai J, Bahi R, et al. Are urologists and radi-
ologists equally effective in determining the
RENAL Nephrometry score? Ann Surg On-
col. 2015 May;22(5):1618–24.
18 Gu L, Ma X, Li H, Yao Y, Xie Y, Chen L, et al.
External validation of the Arterial Based
Complexity (ABC) scoring system in renal tu-
mors treated by minimally invasive partial ne-
phrectomy. J Surg Oncol. 2017 Sep; 116(4):
507–14.
19 Roscigno M, Ceresoli F, Naspro R, Montorsi
F, Bertini R, Da Pozzo LF. Predictive accuracy
of nephrometric scores can be improved by
adding clinical patient characteristics: a novel
algorithm combining anatomic tumour com-
plexity, body mass index, and Charlson co-
morbidity index to depict perioperative com-
plications after nephron-sparing surgery. Eur
Urol. 2014 Jan;65(1):259–62.
20 Kriegmair MC, Mandel P, Moses A, Lenk J,
Rothamel M, Budjan J, et al. Defining Renal
Masses: Comprehensive Comparison of RE-
NAL, PADUA, NePhRO, and C-Index Score.
Clin Genitourin Cancer. 2017 Apr;15(2):248–
255.e1.
21 Ljungberg B, Albiges L, Abu-Ghanem Y, Ben-
salah K, Dabestani S, Fernández-Pello S, et al.
European Association of Urology Guidelines
on Renal Cell Carcinoma: The 2019 Update.
Eur Urol. 2019 May;75(5):799–810.
130.209.6.61 - 11/30/2019 10:20:56 AM

Predictive Kidney Models for Surgery Urol Int 7

Type DOI: 10.1159/000504145
Glasgow Univ.Lib.
Downloaded by: