Upper Urinary Tract
Development and internal validation of a nomogram for predicting stone-free status after flexible ureteroscopy for renal stones Hiroki Ito*†, Kentaro Sakamaki‡, Takashi Kawahara*†, Hideyuki Terao*, Kengo Yasuda†, Shinnosuke Kuroda†, Masahiro Yao*, Yoshinobu Kubota* and Junichi Matsuzaki† *Department of Urology, Yokohama City University Graduate School of Medicine, †Department of Urology, Ohguchi East General Hospital, and ‡Department of Biostatistics and Epidemiology, Yokohama City University Graduate School of Medicine, Yokohama, Japan
Objective To develop and internally validate a preoperative nomogram for predicting stone-free status (SF) after ﬂexible ureteroscopy (fURS) for renal stones, as there is a need to predict the outcome of fURS for the treatment of renal stone disease.
Patients and Methods
(P = 0.001), operator with experience of >50 fURS (P = 0.026), stone number (P = 0.075), and presence of hydronephrosis (P = 0.047). We developed a nomogram to predict SF after fURS using these ﬁve preoperative characteristics. Total nomogram score (maximum 25) was derived from summing individual scores of each predictive variable; a high total score was predictive of successful fURS outcome, whereas a low total score was predictive of unsuccessful outcome. The area under the receiver operating characteristics for nomogram predictions was 0.87.
We retrospectively analysed 310 fURS procedures for renal stone removal performed between December 2009 and April 2013. Final outcome of fURS was determined by computed tomography 3 months after the last fURS session. Assessed preoperative factors included stone volume and number, age, sex, presence of hydronephrosis and lower pole calculi, and ureteric stent placement. Multivariate logistic regression analysis with backward selection was used to model the relationship between preoperative factors and SF after fURS. Bootstrapping was used to internally validate the nomogram.
Five independent predictors of SF after fURS were identiﬁed: stone volume (P < 0.001), presence of lower pole calculi
ureteroscopy, kidney stone, lithotripsy
eﬀective for complex intrarenal stone removal [3,4,6,10]. Selection of the optimal procedure for renal stone clearance is now more complex because of the introduction of fURS, and there is consequently a need for methods that predict the success rate of URS.
Both the European Association of Urology (EAU) and the AUA clinical guidelines do not currently recommend ﬂexible ureteroscopy (fURS) as a ﬁrst-line procedural treatment for renal stones [1,2]. However, URS has become a safer and more accepted method for treating renal stones with the development and use of ﬂexible ureteroscopes [3–8]. Several reports showed that fURS is a suitable alternative to shockwave lithotripsy (SWL) and percutaneous nephrolithotripsy (PCNL) for the treatment of renal stones, with potentially higher stone-free rates than SWL and lower morbidity than PCNL [4,5,7–9]. Safe and eﬀective removal by fURS of multiple and large intrarenal stones. including lower pole calculi. has been documented, suggesting that fURS is
BJU Int 2015; 115: 446–451 wileyonlinelibrary.com
The nomogram can be used to reliably predict SF based on patient characteristics after fURS treatment of renal stone disease.
Established models for predicting the outcomes of SWL and PCNL exist [11–14] but a nomogram for predicting the outcome of URS is still in development , particularly for fURS treatment of renal stones. In previous studies, we demonstrated that stone volume is the most signiﬁcant factor predictive of stone-free status (SF) after fURS [16,17]. We speculate that stone volume combined with other variables might improve prediction of SF after fURS. The aim of the present study was to develop and internally validate a © 2014 The Authors BJU International © 2014 BJU International | doi:10.1111/bju.12775 Published by John Wiley & Sons Ltd. www.bjui.org
Nomogram for predicting stone-free status after fURS
preoperative nomogram for predicting SF after fURS treatment of renal stone disease.
Patients and Methods We retrospectively analysed 388 fURS for renal stone removal performed between December 2009 and April 2013 at Ohguchi East General Hospital. Of these 388 procedures, 78 were excluded from the study due to the following factors: lack of analysed parameters noted, presence of staghorn calculi or sponge kidney, or more than two stages of URS in multistage fURS. A staghorn calculi was deﬁned as involving the renal pelvis and all renal calyces in the present study. The remaining 310 procedures consisted of ﬁrst-time URS or ﬁrst-stage fURS in multistage fURS. Our approach to the treatment of urinary stones was as follows: SWL and PCNL were recommended for patients with urinary stones of 20 mm in diameter, respectively, while fURS was oﬀered as either a ﬁrst or second option. The ﬁnal decision was made based on both the patient’s and the surgeon’s preference [16,18]. This study was approved by the Institutional Ethics Committee of Ohguchi East General Hospital. We obtained written informed consent from all patients for their data to be used for research purposes. Surgical Techniques Details of our surgical procedures have been previously described [16,18]. Brieﬂy, fURS was performed with a 6/7.5 or 8/9.8 F semi-rigid ureteroscope (WolfTM, Knittlingen, Germany) and/or a 6 F ﬂexible ureteroscope (Flex-X2TM, STORZ, Germany or Olympus P-5TM, Olympus, Tokyo, Japan) with 200–550-μm holmium:yttrium-aluminum-garnet laser lithotripsy. Ureteric access sheaths (12/14 or 14/16 F, Cook Medical, Bloomington, IN or 11/13 or 13/15 F, Boston Scientiﬁc, Natick, MA, USA) were placed to facilitate stone extraction and reduce the intrarenal pressure in cases of renal and ureteric stones except U3 (lower ureter) stones. In all cases, 1.5 F and/or 2.2 F tipless nitinol baskets were used for stone removal and clearance of residual fragments. To minimise perioperative complications, the procedure was stopped if 120 min elapsed since the start of the procedure. Preoperative and Postoperative Evaluation Stone status was evaluated regularly by examining kidney-ureter-bladder X-ray (KUB) ﬁlms beginning on postoperative day 1 (POD1). Stone status at POD1 was determined from intraoperative endoscopic ﬁndings and the KUB ﬁlm at POD1. The ﬁnal outcome of fURS was assessed 3 months after the last fURS session by non-contrast CT (NCCT). SF was deﬁned as the strict absence of visible stones on imaging.
Assessed preoperative factors include stone diameter (mm), volume (mm3), and number, side of involvement (right or left), age, sex, height, body weight, body mass index (BMI), presence of hydronephrosis (including partial hydronephrosis) or lower pole calculi, and placement of ureteric stent. Stone volume was measured using 5-mm axial and 3.5-mm reconstructed coronal NCCT images as previously reported [16,18]. Stone number and location and the presence of hydronephrosis or lower pole calculi were evaluated by preoperative NCCT. Stone status was assessed by the same urologist (H.I.) and the procedures were performed by a total of eight urologists. Statistical Analysis Patient characteristics and preoperative factors were analysed using Mann–Whitney U and chi-square tests. Continuous variables are expressed as the mean (SD). Statistical signiﬁcance was set at P < 0.05. Multivariate logistic regression models were used to assess the success rate of fURS. Reduced model selection was performed using backward selection. The selection criterion of P < 0.1 was used for elimination of a variable. Some variables were initially excluded from the multivariate model because of multicollinearity. Decisions with respect to the coding of the nomogram variables were made before variable selection. Regression coeﬃcients were used to generate prognostic nomograms. The predictive ability of the nomogram was evaluated by the area under the receiver operating characteristics (AUROC) curve. Internal validation was performed using 1000 bootstrap resamples. A bootstrapping method is a nonparametric data generating method in which new datasets are repeatedly generated from an original dataset. Therefore, it is used for internal validation, not external validation. Statistical analyses were performed using SAS 9.3 (SAS Institute Inc., Cary, NC, USA).
Results Patient Characteristics and Surgical Outcome Of the 310 procedural cases analysed in this study, 301 were performed under general anaesthesia and nine under spinal anaesthesia. A ureteric sheath was used in all cases. Table 1 compares patient characteristics and treatment outcomes according to SF at 3 months after the procedure. Signiﬁcant diﬀerences between the SF and non-SF groups were seen for the following parameters: stone number (P < 0.001), stone burden including stone diameter and volume (P < 0.001), presence of lower pole calculi (P < 0.001), maximum and mean Hounsﬁeld unit (P < 0.001), duration of procedure (P < 0.001), energy of laser (P < 0.001), and postoperative duration of catheter use (P = 0.001). There were no © 2014 The Authors BJU International © 2014 BJU International
Ito et al.
Table 1 Comparison of patient and stone characteristics according to SF at 3 months after fURS. Variable Number of patients Mean (SD) age, years Sex, n: Female Male Side, n: Right Left Lower pole calculi +/–, n Hydronephrosis +/–, n Preoperative stenting, n (emergent/intentional) Mean (SD) Height, cm Body weight, kg BMI, kg/m2 Stone number Stone burden Diameter, mm Volume, mm3 Hounsﬁeld units: Maximum Mean Extracorporeal SWL failure, n Mean (SD) duration of procedure, min Mean (SD) energy of laser, kJ Postoperative catheter duration 1 day/1–2 months, n Experience of operator