Split Renal Function Is Fundamentally Important for Predicting Functional Recovery After Radical Nephrectomy

Automated Analysis of Split Kidney Function from CT Scans Using Deep Learning and Delta Radiomics

Background: Differential kidney function assessment is an important part of preoperative evaluation of various urological interventions. It is obtained through dedicated nuclear medical imaging and is not yet implemented through conventional Imaging. Objective: We assess if differential kidney function can be obtained through evaluation of contrast-enhanced computed tomography(CT) using a combination of deep learning and (2D and 3D) radiomic features. Methods: All patients who underwent kidney nuclear scanning at Mayo Clinic sites between 2018-2022 were collected. CT scans of the kidneys were obtained within a 3-month interval before or after the nuclear scans were extracted. Patients who underwent a urological or radiological intervention within this time frame were excluded. A segmentation model was used to segment both kidneys. 2D and 3D radiomics features were extracted and compared between the two kidneys to compute delta radiomics and assess its ability to predict differential kidney function. Performance was reported using receiver operating characteristics, sensitivity, and specificity. Results: Studies from Arizona & Rochester formed our internal dataset (n = 1,159). Studies from Florida were separately processed as an external test set to validate generalizability. We obtained 323 studies from our internal sites and 39 studies from external sites. The best results were obtained by a random forest model trained on 3D delta radiomics features. This model achieved an area under curve (AUC) of 0.85 and 0.81 on internal and external test sets, while specificity and sensitivity were 0.84,0.68 on the internal set, 0.70, and 0.65 on the external set. Conclusion: This proposed automated pipeline can derive important differential kidney function information from contrast-enhanced CT and reduce the need for dedicated nuclear scans for early-stage differential kidney functional assessment. Clinical Impact: We establish a machine learning methodology for assessing differential kidney function from routine CT without the need for expensive and radioactive nuclear medicine scans.

Practical Prediction of New Baseline Renal Function After Partial Nephrectomy

BACKGROUND

Partial nephrectomy (PN) is generally preferred for localized renal masses due to strong functional outcomes. Accurate prediction of new baseline glomerular filtration rate (NBGFR) after PN may facilitate preoperative counseling because NBGFR may affect long-term survival, particularly for patients with preoperative chronic kidney disease. Methods for predicting parenchymal volume preservation, and by extension NBGFR, have been proposed, including those based on contact surface area (CSA) or direct measurement of tissue likely to be excised/devascularized during PN. We previously reported that presuming 89% of global GFR preservation (the median value saved from previous, independent analyses) is as accurate as the more subjective/labor-intensive CSA and direct measurement approaches. More recently, several promising complex/multivariable predictive algorithms have been published, which typically include tumor, patient, and surgical factors. In this study, we compare our conceptually simple approach (NBGFRPost-PN = 0.90 × GFRPre-PN) with these sophisticated algorithms, presuming that an even 90% of the global GFR is saved with each PN.

PATIENTS AND METHODS

A total of 631 patients with bilateral kidneys who underwent PN at Cleveland Clinic (2012-2014) for localized renal masses with available preoperative/postoperative GFR were analyzed. NBGFR was defined as the final GFR 3-12 months post-PN. Predictive accuracies were assessed from correlation coefficients (r) and mean squared errors (MSE).

RESULTS

Our conceptually simple approach based on uniform 90% functional preservation had equivalent r values when compared with complex, multivariable models, and had the lowest degree of error when predicting NBGFR post-PN.

CONCLUSIONS

Our simple formula performs equally well as complex algorithms when predicting NBGFR after PN. Strong anchoring by preoperative GFR and minimal functional loss (≈ 10%) with the typical PN likely account for these observations. This formula is practical and can facilitate counseling about expected postoperative functional outcomes after PN.

Real-time estimation of nephron activity with a linear measurement system (RENAL-MS) predicts postoperative estimated glomerular filtration rate

OBJECTIVE

To determine whether a simple point-of-care measurement system estimating renal parenchymal volume using tools ubiquitously available could be used to replace nuclear medicine renal scintigraphy (NMRS) in current clinical practice to predict estimated glomerular filtration rate (eGFR) after nephrectomy by estimating preoperative split renal function.

PATIENTS AND METHODS

We performed a retrospective review of patients who underwent abdominal cross-sectional imaging (computed tomography/magnetic resonance imaging) and mercaptoacetyltriglycine (MAG3) NMRS prior to total nephrectomy at a single institution. We developed the real-time estimation of nephron activity with a linear measurement system (RENAL-MS) method of estimating postoperative renal function via the following technique: renal parenchymal volume of the removed kidney relative to the remaining kidney was estimated as the product of renal length and the average of six renal parenchymal thickness measurements. The utility of this value was compared to the utility of the split renal function measured by MAG3 for prediction of eGFR and new onset Stage 3 chronic kidney disease (CKD) at ≥90 days after nephrectomy using uni- and multivariate linear and logistic regression.

RESULTS

A total of 57 patients met the study criteria. The median (interquartile range [IQR]) age was 69 (61-80) years. The median (IQR) pre- and postoperative eGFR was 74 (IQR 58-90) and 46 (35-62) mL/min/1.73 m2 , respectively. [Correction added on 29 December 2023, after first online publication: The data numbers in the preceding sentence have been corrected.] Correlations between actual and predicted postoperative eGFR were similar whether the RENAL-MS or NMRS methods were used, with correlation using RENAL-MS being slightly numerically but not statistically superior (R = 0.82 and 0.76; P = 0.138). Receiver operating characteristic curve analysis using logistic regression estimates incorporating age, sex, and preoperative creatinine to predict postoperative Stage 3 CKD were similar between RENAL-MS and NMRS (area under the curve 0.93 vs. 0.97). [Correction added on 29 December 2023, after first online publication: The data numbers in the preceding sentence have been corrected.] CONCLUSION: A point-of-care tool to estimate renal parenchymal volume (RENAL-MS) performed equally as well as NMRS to predict postoperative eGFR and de novo Stage 3 CKD after nephrectomy in our population, suggesting NMRS may not be necessary in this setting.

Predicting new-baseline glomerular filtration rate (NBGFR) after donor nephrectomy: validation of a split renal function (SRF)-based formula

BACKGROUND

Accurate prediction of post-donor nephrectomy (DN) glomerular filtration rate is potentially useful for evaluating and counselling living kidney donors. Currently, there are limited tools to evaluate post-operative new-baseline glomerular filtration rate (NBGFR) in kidney donors. We aim to validate a conceptually simple formula based on split renal function (SRF) previously developed for radical nephrectomy patients.

METHODS

Eighty-three consecutive patients who underwent DN from 2010 to 2016 were included. Pre-operative CT imaging and functional data including pre-DN baseline Global GFR (108.2 ± 13.2 mL/min/1.73m2) were included. Observed NBGFR was defined as the latest eGFR 3-12 months post-DN. SRF, defined as volume of the contralateral non-resected kidney normalised by total volume of kidneys, was determined from pre-operative cross-sectional imaging (49.2 ± 2.36%). The equation derived from Rathi et al. is as detailed: Predicted NBGFR = 1.24 × (Global GFR Pre-DN) x (SRF).

RESULTS

The relationship between predicted NBGFR (66.0 ± 8.29 mL/min/1.73m2) and observed NBGFR (74.9 ± 16.4 mL/min/1.73m2) was assessed by evaluating correlation coefficients, bias, precision, accuracy, and concordance. The new SRF-based formula for NBGFR prediction correlated strongly with observed post-operative NBGFR (Pearson's r = 0.729) demonstrating minimal bias (median difference = 7.190 mL/min/1.73m2) with good accuracy (96.4% within ± 30%, 62.7% within ± 15%) and precision (IQR of bias =  - 0.094 to 16.227).

CONCLUSION

The SRF-based formula was also able to accurately discriminate all but one patient to an NBGFR of > 45 mL/min/1.73m2. We utilised the newly developed SRF-based formula for predicting NBGFR in a living kidney donor population. Counselling of donor post-operative renal outcomes may then be optimised pre-operatively.

Predicting estimated glomerular filtration rate after partial and radical nephrectomy based on split renal function measured by radionuclide: a large-scale retrospective study

PURPOSE

The purpose of this study was to develop predictive models for postoperative estimated glomerular filtration rate (eGFR) based on the split glomerular filtration rate measured by radionuclide (rGFR), as choosing radical nephrectomy (RN) or partial nephrectomy (PN) for complex renal masses requires accurate prediction of postoperative eGFR.

METHODS

Patients who underwent RN or PN for a single renal mass at Xijing Hospital between 2008 and 2022 were retrospectively included. Preoperative split rGFR was evaluated using technetium-99 m-diethylenetriaminepentaacetic acid (Tc-99 m DTPA) renal dynamic imaging, and the postoperative short-term (< 7 days) and long-term (3 months to 5 years) eGFRs were assessed. Linear mixed-effect models were used to predict eGFRs, with marginal R2 reflecting predictive ability.

RESULTS

After excluding patients with missing follow-up eGFRs, the data of 2251 (RN: 1286, PN: 965) and 2447 (RN: 1417, PN: 1030) patients were respectively included in the long-term and short-term models. Two models were established to predict long-term eGFRs after RN (marginal R2 = 0.554) and PN (marginal R2 = 0.630), respectively. Two other models were established to predict short-term eGFRs after RN (marginal R2 = 0.692) and PN (marginal R2 = 0.656), respectively. In terms of long-term eGFRs, laparoscopic and robotic surgery were superior to open surgery in both PN and RN.

CONCLUSIONS

We developed novel tools for predicting short-term and long-term eGFRs after RN and PN based on split rGFR that can help in preoperative decision-making.

Development of a novel score (RENSAFE) to determine probability of acute kidney injury and renal functional decline post surgery: A multicenter analysis

OBJECTIVE

To create and validate 2 models called RENSAFE (RENalSAFEty) to predict postoperative acute kidney injury (AKI) and development of chronic kidney disease (CKD) stage 3b in patients undergoing partial (PN) or radical nephrectomy (RN) for kidney cancer.

METHODS

Primary objective was to develop a predictive model for AKI (reduction >25% of preoperative eGFR) and de novo CKD≥3b (<45 ml/min/1.73m2), through stepwise logistic regression. Secondary outcomes include elucidation of the relationship between AKI and de novo CKD≥3a (<60 ml/min/1.73m2). Accuracy was tested with receiver operator characteristic area under the curve (AUC).

RESULTS

AKI occurred in 452/1,517 patients (29.8%) and CKD≥3b in 116/903 patients (12.8%). Logistic regression demonstrated male sex (OR = 1.3, P = 0.02), ASA score (OR = 1.3, P < 0.01), hypertension (OR = 1.6, P < 0.001), R.E.N.A.L. score (OR = 1.2, P < 0.001), preoperative eGFR<60 (OR = 1.8, P = 0.009), and RN (OR = 10.4, P < 0.0001) as predictors for AKI. Age (OR 1.0, P < 0.001), diabetes mellitus (OR 2.5, P < 0.001), preoperative eGFR <60 (OR 3.6, P < 0.001) and RN (OR 2.2, P < 0.01) were predictors for CKD≥3b. AUC for RENSAFE AKI was 0.80 and 0.76 for CKD≥3b. AKI was predictive for CKD≥3a (OR = 2.2, P < 0.001), but not CKD≥3b (P = 0.1). Using 21% threshold probability for AKI achieved sensitivity: 80.3%, specificity: 61.7% and negative predictive value (NPV): 88.1%. Using 8% cutoff for CKD≥3b achieved sensitivity: 75%, specificity: 65.7%, and NPV: 96%.

CONCLUSION

RENSAFE models utilizing perioperative variables that can predict AKI and CKD may help guide shared decision making. Impact of postsurgical AKI was limited to less severe CKD (eGFR<60 ml/min 71.73m2). Confirmatory studies are requisite.

Long-Term Renal Function Following Renal Cancer Surgery: Historical Perspectives, Current Status, and Future Considerations

Knowledge of functional recovery after partial (PN) and radical nephrectomy for renal cancer has advanced considerably, with PN now established as the reference standard for most localized renal masses. However, it is still unclear whether PN provides an overall survival benefit in patients with a normal contralateral kidney. While early studies seemingly demonstrated the importance of minimizing warm-ischemia time during PN, multiple new investigations over the last 10 years have proven that parenchymal mass lost is the most important predictor of new baseline renal function. Minimizing loss of parenchymal mass during resection and reconstruction is the most important controllable aspect of long-term post-operative renal function preservation.

Point of care parenchymal volume analyses to estimate split renal function and predict functional outcomes after radical nephrectomy

Accurate prediction of new baseline GFR (NBGFR) after radical nephrectomy (RN) can inform clinical management and patient counseling whenever RN is a strong consideration. Preoperative global GFR, split renal function (SRF), and renal functional compensation (RFC) are fundamentally important for the accurate prediction of NBGFR post-RN. While SRF has traditionally been obtained from nuclear renal scans (NRS), differential parenchymal volume analysis (PVA) via software analysis may be more accurate. A simplified approach to estimate parenchymal volumes and SRF based on length/width/height measurements (LWH) has also been proposed. We compare the accuracies of these three methods for determining SRF, and, by extension, predicting NBGFR after RN. All 235 renal cancer patients managed with RN (2006-2021) with available preoperative CT/MRI and NRS, and relevant functional data were analyzed. PVA was performed on CT/MRI using semi-automated software, and LWH measurements were obtained from CT/MRI images. RFC was presumed to be 25%, and thus: Predicted NBGFR = 1.25 × Global GFRPre-RN × SRFContralateral. Predictive accuracies were assessed by mean squared error (MSE) and correlation coefficients (r). The r values for the LWH/NRS/software-derived PVA approaches were 0.72/0.71/0.86, respectively (p < 0.05). The PVA-based approach also had the most favorable MSE, which were 120/126/65, respectively (p < 0.05). Our data show that software-derived PVA provides more accurate and precise SRF estimations and predictions of NBGFR post-RN than NRS/LWH methods. Furthermore, the LWH approach is equivalent to NRS, precluding the need for NRS in most patients.

Partial Versus Radical Nephrectomy: Complexity of Decision-Making and Utility of AUA Guidelines

INTRODUCTION

The American-Urological-Association(AUA) Guidelines for renal cancer(2017) recommend consideration for radical-nephrectomy(RN) over partial(PN) whenever there is increased oncologic-risk; and RN should be prioritized if three other criteria are all also met: 1) increased tumor-complexity; 2) no preexisting chronic-kidney-disease/ proteinuria, and 3) normal contralateral kidney that will likely provide estimated glomerular-filtration-rate (eGFR) >45ml/min/1.73m² even if RN is performed. Our objective was to assess the complexity of decision-making about RN/PN and utility of AUA Guidelines statements regarding this issue.

PATIENTS AND METHODS

Retrospective review of 267 consecutive RN/PN from 2019(100-RN/167-PN). High tumor-complexity was defined as R.E.N.A.L.≥9. Increased oncologic-risk was defined as tumor >7cm, locally-advanced or infiltrative-features on imaging, or high-risk pathology on biopsy, if obtained. New-baseline GFR after RN was estimated using global-GFR, split-renal-function_{contralateral}, and presuming 25% renal-functional-compensation.

RESULTS

163 patients(61%) fit scenarios that are well-defined in the Guidelines. Of these, 34 had strong indications for RN, and all had RN. Twelve of 129 patients(9.3%) underwent RN despite Guidelines generally favoring PN. The remaining 104 patients(39%) did not fit within situations where the Guidelines provide specific recommendations. In these patients, RN was often performed despite functional-considerations favoring PN due to overriding concerns about oncologic-risk and/or tumor-complexity.

CONCLUSION

Our data demonstrate complexity of decision-making about PN/RN as almost 40% of patients did not fit well-described AUA Guidelines descriptors. Compliance was generally strong although occasional overutilization of RN remains a concern in our series, and will be addressed with additional education. Further studies will be required to assess the generalizability of our findings in other institutions/settings.