ESPU Meeting on Thursday 9, June 2022, 09:15 - 10:10
09:15 - 09:18
S05-1 (OP)
Sang Hoon SONG 1, Jae Hyeon HAN 2, Kun Suk KIM 1, Young Ah CHO 3, Hyejung YOUN 4, Young In KIM 5 and Jihoon KWEON 4
1) Asan Medical Center, University of Ulsan College of Medicine, Department of Urology, Seoul, REPUBLIC OF KOREA - 2) Korea University Ansan Hospital, Korea University College of Medicine, Seoul, Republic of Korea, Department of Urology, Seoul, REPUBLIC OF KOREA - 3) Asan Medical Center, University of Ulsan College of Medicine, Department of Radiology and Research Institute of Radiology, Seoul, REPUBLIC OF KOREA - 4) Asan Medical Center, Department of Convergence Medicine, Seoul, REPUBLIC OF KOREA - 5) Asan Medical Institute of Convergence Science and Technology, Asan Medical Center, Department of Medical Science, Seoul, REPUBLIC OF KOREA
PURPOSE
We investigated the feasibility of measuring the hydronephrosis area to renal parenchyma (HARP) ratio from ultrasound images using a deep-learning network.
MATERIAL AND METHODS
The coronal renal ultrasound images of 195 pediatric and adolescent patients who underwent pyeloplasty due to ureteropelvic junction obstruction were retrospectively reviewed. After excluding cases without the representative longitudinal renal image, we used a dataset of 168 images for deep-learning segmentation. Ten novel networks, such as combinations of DeepLabV3+ and UNet++ were assessed for hydronephrosis (HN) and kidney areas, and the ensemble method was applied for further improvement. By dividing the image set into four, cross-validation was conducted, and the segmentation performance of the deep-learning network was evaluated using sensitivity, specificity, and dice similarity coefficient by comparing it with the manually traced area.
RESULTS
All 10 networks and ensemble methods showed a good visual correlation with the manually traced kidney and HN areas. The dice similarity coefficient of the 10-model ensemble was 0.9108 on average, and the best 5-model ensemble had a dice similarity coefficient of 0.9113 on average. We included patients with severe HN who underwent renal US examination in a single institution; thus, external validation of our algorithm in a heterogeneous US examination setup with a diverse set of US instruments is recommended.
CONCLUSIONS
Deep learning-based calculation of the HARP ratio is feasible and showed high performance with accuracy for imaging of the severity of HN using ultrasonography. This algorithm can help physicians make a more accurate and reproductive imaging diagnosis of HN using US.
09:18 - 09:21
S05-2 (OP)
Roseanne FERREIRA DE FREITAS EUZEBIO 1, Melissa MCGRATH 1, Bruno LESLIE 1 and Luis H. BRAGA 2
1) McMaster University, Surgery-Urology, Hamilton, CANADA - 2) McMaster University - McMaster Children's Hospital, Department of Surgery / Urology, Hamilton, CANADA
PURPOSE
To compare the ability of the antero-posterior pelvic diameter (APD) vs. Urinary tract dilation (UTD) grading systems to predict the rate and time to resolution of hydronephrosis (HN), according to its severity.
MATERIAL AND METHODS
A prospectively collected cohort of patients with prenatal HN was followed from 2008-2021.
We excluded patients with hydroureteronephrosis, vesicoureteral reflux and other urinary tract anomalies. Patients who underwent surgery were censored as non-resolved HN at that time. Each HN grading system was dichotomized for comparison purpose: APD <15 vs. ³15mm and UTD P1 vs. UTD P2/P3. HN resolution was defined as APD <5mm or UTD P0 in 2 consecutive ultrasounds. Kaplan-Meier, Cox regression and weighted Kappa were used for statistical analysis.
RESULTS
Of 518 patients, 243 (47%) were classified as UTD P1, 192 (37%) as UTD P2 and 82 (16%) as UTD P3. Mean APD and age at baseline were 11.6±7.5mm and 3.2±2.5 months, respectively. A total of 188 (36%) patients resolved at a median time of 32 months. Analysis of agreement showed 362 of 515 (70%) infants were classified as similar HN severity by both grading systems, and 153 (30%) were UTD P2 or P3 with APD <15mm, K=0.42, p<0.001 (table 2). Median time to resolution for different HN grades is presented in table 1. The overall cumulative resolution rate was 82% and 60% for <15mm and ³15mm APD, respectively. UTD’s cumulative resolution rates were 85% for UTD P1 and 71% for UTD P2/P3. (figure 1).
CONCLUSIONS
Time to resolution of HN was similar regardless of the grading system used. Both APD and UTD classifications can be effectively applied clinically to counsel patients about resolution trends.
09:21 - 09:24
S05-3 (OP)
Ruby WILLIAMS 1, Harriet CORBETT 2 and Umber AGARWAL 3
1) Alder Hey Children's NHS Foundation Trust, Department of Surgery, Liverpool, UNITED KINGDOM - 2) Alder Hey Children's NHS Foundation Trust, Regional Department of Urology, Liverpool, UNITED KINGDOM - 3) Liverpool Women's Hospital, Fetal Medicine Department, Liverpool, UNITED KINGDOM
PURPOSE
To identify when obstetricians would deliver a fetus with antenatal hydronephrosis and normal liquor.
MATERIAL AND METHODS
The study design was a Snap-shot survey, the survey was accessed via a Survey Monkey link. The survey was aimed at obstetrics and fetal medicine consultants and the survey link was distributed via closed professional forums on the UK North West Coast Maternity Clinical Network, Facebook, and publicly on Twitter. Respondents were asked the number of years they had been in practice as a consultant, at what gestation would they deliver a fetus with antenatal hydronephrosis and normal liquor and what criteria they use to make that decision.
RESULTS
44/102 respondents (43%) would deliver prior to 40 weeks (median no of years as consultant 10 years (IQR 5-17) versus those who would not, median years as consultant 5.5 (3-12). Re APD threshold of delivery: 17 indicated delivery if the APD were 20mm, 10 if it were 21-30mm and 16 if it were >30mm. Re gestation at which they would deliver: 13 indicated 37-38weeks, 13 indicated 38-39 weeks and 17 indicated 39-40 weeks. Reasons selected for delivery before term were obstetric anxiety n=2, maternal request n=2, maternal anxiety n=2 and concern about fetal renal damage / renal n=34.
Additional comments followed themes:
[1] Decisions should be made in conjunction with Paediatric Urology team
[2] APD alone is not sensitive enough to make the decision
[3] Concern at the lack of evidence-based practice
[4] There is a need for regional/national guidelines
CONCLUSIONS
A surprising number of respondents would consider early delivery of a fetus with hydronephrosis and normal liquor despite the lack of evidence of benefit. Data regarding neurodevelopmental outcomes and socioeconomic achievement supports term (rather than pre- or early term) delivery in this scenario unless there are obstetric reasons for early delivery.
09:24 - 09:27
S05-4 (OP)
Fayez ALMODHEN 1, Wael MONEIR 1, Abdullah AL-AQEEL 1, Abdulwahab AL-HAMS 1, Ahmad AL-SHAMMARI 1, Yasser JAMALALAIL 1, Tariq BURKI 1, Ahmad AL-ZAHRANI 2 and Ameen AMEEN BASHAREEF 1
1) King Abdulaziz Medical City, Section of Pediatric Urology. Department of Pediatric Surgery, Riyadh, SAUDI ARABIA - 2) King Abdulaziz Medical City, Department of Radiology, Riyadh, SAUDI ARABIA
PURPOSE
Although ureteropelvic junction obstruction (UPJO) remains the main surgical category of antenatally detected hydronephrosis, there is a lack of a gold standard test that predicts the need for surgery. We highlighted a novel parameter, the calyceal-to-parenchymal ratio (CPR) of postnatal renal ultrasonography (RUS) as a predictor of surgery in newborns with possible obstructive hydronephrosis
MATERIAL AND METHODS
We identified 79 renal units that were managed with pyeloplasty between January 2010 and July 2020, and 79 randomly selected comparable renal units managed conservatively were included in the control group. We compared the CPR of the postnatal RUS between the two groups, and tested the correlation between CPR and the results of the diuretic renogram
RESULTS
At a mean age of 18.9 weeks and a mean follow-up period of 48.99 months, the median CPR of the postnatal RUS was significantly greater in the surgical group than in the conservative group (3.62 vs. 0.98, p <0.001). A CPR of 1.68 had a sensitivity and specificity of 96.2% and 84.8%, respectively, in predicting the need for pyeloplasty. The data revealed a statistically significant relationship between postnatal CPR and the drainage curve of the renogram (p = 0.018) and the postnatal degree of hydronephrosis (p < 0.001). Postnatal CPR ranks were significantly higher than postoperative CPR ranks (p < 0.001)
CONCLUSIONS
Postnatal CPR is a promising tool for predicting the need for surgery in patients with UPJO with antenatal hydronephrosis. Further prospective studies are needed to standardize and assess the reproducibility of this parameter
09:41 - 09:44
S05-5 (OP)
Mohamed SOLTAN, Ahmed ABDELHALIM, Mohamed EDWAN, Ahmed ELKASHEF, Ahmed ATWA, Hussam KHALIFA, Yahya HUSSAM, Ahmed ELGHAREEB, Mohamed ABOU-EL_GHAR, Tamer HELMY, Ashraf TAREK, Mohamed DAWABA and Ahmed SHOKIER
Mansoura urology and nephrology centre, Urology, Mansoura, EGYPT
PURPOSE
Resolution of hydronephrosis is still not well understood and time to hydronephrosis resolution varies greatly among subjects. Consequently, postoperative follow up protocols are different worldwide
MATERIAL AND METHODS
Data of children < 15 years who underwent pyeloplasty at a tertiary center between January 2015 and October 2019 were reviewed. patients who underwent redo pyeloplasty and those with missing data were excluded. hydronephrosis resolution was defined as decrease of the APD to less than 10 mm or >50% reduction of the APD. Analysis of survival using Kaplan Meier's curve was performed with adjustment to perioperative and clinical data
RESULTS
A total of 256 children were included in the study.Hydronephrosis resolution rate was 74.6%, 87.3%, 93.7% and 97% at 1, 2, 3 and 4 years of follow-up; respectively. Univariate analysis showed no difference in resolution neither between patients with preoperative APD <2, 2-4 and more than 4 cms (p= 0.15) nor between different hydronephrosis grades (p=0.38). No difference was found among patient age groups <1, 1-5, 5-10 and >10 years old (log rank 0.21). Also bilateral cases and those poorly functioning kidneys (<20% split renal function) did not show any difference in resolution of hydronephrosis (log rank 0.2, and 0.7 respectively. patients with solitary kidneys had less resolution rates (log rank 0.03). In addition, cases with percent of improvement of APD (PIAPD) more than 10 % within the initial visit showed significantly better resolution rates (log rank 0.0001). Both were significant in cox regression analysis.
CONCLUSIONS
In more than 90 % of cases, hydronephrosis tend to resolve within 3 years after pyeloplasty. Patients with decrease in APD more than 10 % from the preoperative value have higher rates of resolution, however patients with solitary kidney have lower rates of resolution.
09:44 - 09:47
S05-6 (OP)
Sonia PÉREZ-BERTÓLEZ 1, Oriol MARTIN-SOLE 2, Isabel CASAL-BELOY 2, Blanca CAPDEVILA 2, Mar CARBONELL 2 and Luis GARCIA-APARICIO 2
1) Hospital Sant Joan de Déu, Pediatric Urology Unit, Department of Pediatric Surgery, Barcelona, SPAIN - 2) Hospital Sant Joan de Déu, Pediatric Urology Unit. Department of Pediatric Urology., Barcelona, SPAIN
PURPOSE
The high-pressure balloon dilatation (HPBD) of the ureterovesical junction with double-J stenting is a minimally invasive alternative for first line primary obstructive megaureter (POM) surgical treatment instead uretearl reimplantation or cutaneous ureterostomy. The aim of our study is to identify the risk factors associated with the need of ureteric reimplantation due to failure of HPBD.
MATERIAL AND METHODS
Prospective data collection of patients who underwent HPBD for POM from 2017-2021 at a single institution. Collected data were: patients' demographics, diagnostic modalities, surgical details, results and follow-up. A multivariate logistic regression analysis was performed in order to identify risk factors for need of secondary ureteric reimplantation.
RESULTS
Fifty-five ureters underwent HPBD for POM 50 children, with a median age of 6.4 months (IQR: 4.5-13.8). Nineteen patients (37.25%) underwent secondary ureteric reimplantation, a median of 9.8 months after primary HBPD (95%CI: 6.2 to 9.9). Median follow-up was 29.4 months (IQR: 17.4 to 71). Independent risk factors for redo-surgery in a multivariate logistic regression model were: progressive ureterohydronephrosis (OR=7.8; 95%CI: 0.77-78.6) and early removal of double-J stent. A risk reduction of 7% (95%CI: 2.2%-11.4%) appears per extra-day of catheter maintenance. The optimal cut-off point is 55 days (it predicts failure with a Sensitivity of 69% and Specificity of 68%, correctly classifying 69% of patients), ROC curve area: 0.77 (95%CI: 0.62-0.92). Sex, distal ureteral diameter, pelvis diameter, dilatation balloon diameter and preoperative differential renal function did not have an impact on the need of reimplantation.
CONCLUSIONS
The use of double-J stent for at least 55 days seems to avoid the need of a secondary procedure. Therefore, we recommend removing the double-J catheter at least 2 months after HBPD.
09:47 - 09:50
S05-7 (OP)
Ijeoma NWACHUKWU 1, Shahid IQBAL 1, Arash TAGHIZADEH 1, Anu PAUL 1, Pankaj MISHRA 1 and Massimo GARRIBOLI 2
1) Evelina London Children's Hospital, Paediatric Urology, London, UNITED KINGDOM - 2) Evelina London Children's Hospital - Guy's and St Thomas NHS Foundation Trust, Paediatric Urology, London, UNITED KINGDOM
PURPOSE
Endoscopic approach is first line for treatment of primary obstructive megaureter (POM). Treatment options include high pressure balloon dilatation (HPBD) and cutting balloon ureterotomy (CBU).
We aimed to compare outcome and complications of HPBD vs CBU.
MATERIAL AND METHODS
Retrospective review of patients who underwent endoscopic treatment of POM from 2013 to 2022. Demographic and outcome measures are in table 1.
RESULTS
43 Renal units were treated in 39 children. 23 were treated with CBU and 20 had HPBD. The commonest indication for intervention was increasing dilation in 22 (51%) patients followed by decrease in function (33%) and infections (12%).
Overall, there was significant improvement in pre and post intervention AP pelvis (p=.00011) and DU (p=.0114) measurements. No difference was identified between HPBD and CBU
Tot renal unit n=43 HPBD n=20 CBU n=23 p
Median Age at Surgery (m) 16 (2-214) 18 (2-168) 16 (5-214) .541
AP Dilatation pre (mm) 21.6 (+/- 11.2) 23.3 (+/- 11.8) 20.1 (+/- 10.8) .356
AP Dilatation post (mm) 15.4 (+/- 12.8) 15.5 (+/- 10.9) 15.4 (+/- 14.2) .834
DU Dilatation pre (mm) 14.4 (+/- 7.7) 14.9 (+/- 5.6) 13.9 (+/- 9.3) .258
DU Dilatation post (mm) 10 (+/- 5.3) 11 (+/- 5.6) 9 (+/- 5.0) .358
Length FU (m) 20.4 (+/- 20.1) 26.6 (+/- 25.8 14.6 (+/- 10.3) .136
MAG 3 pre (%) 41 (13-62) 35 (13-62) 46 (20-60) .019
MAG 3 post (%) 44 (26-63) 41 (26-59) 49 (36-63) .018
Complications 1 (2%) 0 1 (4%) -
Reintervention 5 (12%) 4 (20%) 1 (4%) .017
CONCLUSIONS
Endoscopic treatment of POM is safe and with high success rate, outcomes of HPBD and CBU are comparable
09:50 - 09:53
S05-8 (OP)
Julia RHEINBERGER 1, Margaux DIERICKX 2 and Guy BOGAERT 1
1) UZ Leuven, Urology, Leuven, BELGIUM - 2) KU Leuven, Faculty of Medicine, Leuven, BELGIUM
PURPOSE
The treatment for obstructive megaureter is a ureteral reimplantation, often combined with ureteral tailoring. However, tailoring has its additional risks. Is it really necessary to perform a tailoring, or can we avoid it?
MATERIAL AND METHODS
Since 1999, we have, in the event of a obstructive megaureter, created a super wide submucosal tunnel during a cross trigonal ureteral reimplantation (Cohen technique) to allow that megaureter to enter the tunnel tension free and without the need for tailoring. During postoperative follow-up, we specifically focused on the regression of ureter dilatation by at least 30% measured by repeated ultrasounds. Exclusion criteria were previous surgery on kidney, ureter or urethra.
RESULTS
We were able to include and analyze 47 children (mean age at surgery 24 months) with a obstructive megaureter. The mean juxtavesical diameter of the obstructive megaureter was 12 mm measured by ultrasound. Three months postoperatively we observed a decrease in ureteral diameter of at least 30% in 84% of the reimplantations. At 22 months of follow-up, only 7% did not reach a decrease of ureteral diameter. None of the children developed a postoperative ureteral obstruction, however 1/47 children developed vesicoureteral reflux.
CONCLUSIONS
In this study, we were able to demonstrate that, in the event of being able to create a super wide submucosal tunnel allowing tension free cross trigonal reimplantation, tailoring of the megaureter can be safely avoided. Our study shows that a decrease of 30% of the preoperative diameter is to be expected within the first postoperative year.