Best ultrasound parameter for prediction of adverse renal function outcome after pyeloplasty

Sonographic follow-up after pyeloplasty: a large, retrospective cohort analysis

PURPOSE

Routine scintigraphy after surgery for uretero-pelvic junction obstruction (UPJO) is discouraged, making ultrasound the preferred option for follow up. Yet, interpretation of sonographic parameters is rarely straightforward.

METHODS

We reviewed 111 cases including 97 pyeloplasty (52 open, 45 laparoscopic) and 14 pyelopexy during a 7-year period. Pre- and postoperative pelvic antero-posterior diameter (APD), cortical thickness (CT) and pelvis/cortex ratio (PCR) was measured serially.

RESULTS

85% were free of symptoms by 1 year. Only 11% had complete resolution of hydronephrosis. Eleven (10.4%) needed a redo procedure. Mean reduction in APD was 32.6%, 45.8%, and 51.7% at 6 weeks, 3 and 6 months respectively. CT increased by an average 55.9%, 75.6% and 107.6% while PCR reduced by 6.9, 8.0 and 8.8 at given intervals. Comparison of open and laparoscopic procedures showed no significant difference. Review of failed pyeloplasty showed failure of reduction in APD (APD > 3 cm or < 25% reduction) and PCR (PCR > 4) as early indicators for failure.

CONCLUSION

Both APD and PCR are reliable indicators of success and failure following pyeloplasty while CT alone is not as useful. Laparoscopic procedures are non-inferior to standard open surgery.

Deep-learning segmentation of ultrasound images for automated calculation of the hydronephrosis area to renal parenchyma ratio

Purpose

We investigated the feasibility of measuring the hydronephrosis area to renal parenchyma (HARP) ratio from ultrasound images using a deep-learning network.

Materials and Methods

The coronal renal ultrasound images of 195 pediatric and adolescent patients who underwent pyeloplasty to repair ureteropelvic junction obstruction were retrospectively reviewed. After excluding cases without a 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 their ability to calculate hydronephrosis 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 coefficients by comparison with the manually traced area.

Results

All 10 networks and ensemble methods showed good visual correlation with the manually traced kidney and hydronephrosis 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 hydronephrosis who underwent renal ultrasonography at a single institution; thus, external validation of our algorithm in a heterogeneous ultrasonography examination setup with a diverse set of instruments is recommended.

Conclusions

Deep-learning-based calculation of the HARP ratio is feasible and showed high accuracy for imaging of the severity of hydronephrosis using ultrasonography. This algorithm can help physicians make more accurate and reproducible diagnoses of hydronephrosis using ultrasonography.

Renal Cortical Thickness After Pyeloplasty in Pediatric Ureteropelvic Junction Obstruction

Purpose

The main goal of pyeloplasty is to maintain or improve renal function. Diuretic renography is the gold standard for evaluating renal function after pyeloplasty. Renal ultrasonography (RUS) is commonly used to determine hydronephrosis in pediatric patients. We hypothesized that the change in the renal parenchymal cortex would predict pyeloplasty success. In this study, we aim to measure renal cortical thickness change after pyeloplasty in ureteropelvic junction obstruction patients.

Materials and Methods

We retrospectively reviewed 38 patients who underwent pyeloplasty between 2005 and 2019. We divided patients into three age groups and compared the difference of renal parenchymal thickness change by using generalized estimating equations to identify associated factors for renal cortical thickness change after pyeloplasty.

Results

Thirty-nine kidney units were identified. The median age at the time of surgery was 41.61±40.99 months. Generalized estimating equations showed significant change of renal parenchymal thickness over the period of follow-up (p=0.02). The estimate of thickness change was 0.0373 mm/month. Age at the time of surgery was not associated with significant renal cortical thickness over the period of follow-up. The positive predictive factors for renal cortical thickness were creatinine clearance, pre-operative anteroposterior diameter and pre-operative renal differential function. The negative factors for decrease of renal cortical thickness were body weight, presentation with abdominal mass, and history of infection.

Conclusion

Renal cortical thickness after pyeloplasty was improved over the period of follow-up. Age at surgery was not associated with improvement of renal parenchymal thickness.

Pyeloplasty is a safe and effective surgical approach for low functioning kidneys with ureteropelvic junction obstruction

INTRODUCTION

Indications for treatment of ureteropelvic junction obstruction (UPJO) include symptomatic obstruction, urinary tract infections, presence of an obstructive pattern on functional renal scan and/or worsening differential renal function (DRF). This paper aims to determine the relationship between preoperative DRF and surgical outcomes after pyeloplasty. We hypothesized that low preoperative DRF is not an independent predictor of pyeloplasty failure.

METHODS

A retrospective chart review was performed to identify all patients undergoing pyeloplasty for UPJO between 2008 and 2019. Patients were included only if they had at least one preoperative functional scan and a minimum of one renal ultrasound post-operatively. Patients were divided into three groups based on DRF for analysis: Group 1- 0-10%, Group 2 - >10-≤20%, Group 3 - >20%. Baseline, intraoperative and postoperative characteristics, including success and complications were compared. Additional sensitivity analyses were performed comparing patients with ≤20%, and >20% function, ≤30%, and >30% function as well as an analysis of patients undergoing only minimally invasive reconstruction.

RESULTS

Three hundred and sixty-four patients met inclusion criteria. We identified 8 patients in Group 1, 24 patients in Group 2 and 332 patients in Group 3. Mean procedure time was longest for the ≤10% function group (237.9 vs 206.4 vs 189.1; p = 0.01). We found no difference in 30-day post-operative complications, overall success rate or the need for additional procedures among the three groups. For patients in Group 1, we noted variation in the post-procedure DRF with a range of -2.8 to +47% change. In this group, none of patients with low DRF underwent nephrectomy. Multivariate logistic regression did not identify renal function as a predictor of operative success OR 1.00 (95% CI: 0.97-1.03) (p-value: 0.88).

DISCUSSION

The results of the present study suggest that low DRF alone is not associated with worse outcomes and shows no difference in the failure rate. The incidence and type of complications were not increased for the lower functioning groups. The main limitation of this study would be its retrospective nature and single-institution experience. Furthermore, post-operative functional studies were not available for all patients, limiting the ability to draw conclusions on the change in DRF after surgery.

CONCLUSIONS

In a large cohort, preoperative DRF was not predictive of pyeloplasty success rate. DRF ≤10% was not associated with higher incidence of complications or failure rate. The DRF alone should not dictate the management options available for patients with UPJO.