Comparison of RENAL, PADUA, and C-index scoring systems in predicting perioperative outcomes after nephron sparing surgery

Prediction of Postoperative Creatinine Levels by Artificial Intelligence after Partial Nephrectomy

Background and Objectives: Multiple factors are associated with postoperative functional outcomes, such as acute kidney injury (AKI), following partial nephrectomy (PN). The pre-, peri-, and postoperative factors are heavily intertwined and change dynamically, making it difficult to predict postoperative renal function. Therefore, we aimed to build an artificial intelligence (AI) model that utilizes perioperative factors to predict residual renal function and incidence of AKI following PN. Methods and Materials: This retrospective study included 785 patients (training set 706, test set 79) from six tertiary referral centers who underwent open or robotic PN. Forty-four perioperative features were used as inputs to train the AI prediction model. XG-Boost and genetic algorithms were used for the final model selection and to determine feature importance. The primary outcome measure was immediate postoperative serum creatinine (Cr) level. The secondary outcome was the incidence of AKI (estimated glomerular filtration rate (eGFR) < 60 mL/h). The average difference between the true and predicted serum Cr levels was considered the mean absolute error (MAE) and was used as a model evaluation parameter. Results: An AI model for predicting immediate postoperative serum Cr levels was selected from 2000 candidates by providing the lowest MAE (0.03 mg/dL). The model-predicted immediate postoperative serum Cr levels correlated closely with the measured values (R2 = 0.9669). The sensitivity and specificity of the model for predicting AKI were 85.5% and 99.7% in the training set, and 100.0% and 100.0% in the test set, respectively. The limitations of this study included its retrospective design. Conclusions: Our AI model successfully predicted accurate serum Cr levels and the likelihood of AKI. The accuracy of our model suggests that personalized guidelines to optimize multidisciplinary plans involving pre- and postoperative care need to be developed.

Evaluation of modified RENAL nephrometry score in the prediction of perioperative outcomes of open partial nephrectomy

Introduction

RENAL nephrometry score (RNS) is a standardized system to grade the complexity of renal masses, but it does not correlate well with the perioperative outcomes of open partial nephrectomy (OPN). To overcome these shortcomings, a modified RNS (MRNS) has been proposed. In this study, we evaluated the MRNS and its role in predicting the perioperative outcomes of OPN.

Methods

This was a prospective observational study performed at a tertiary care hospital to evaluate the efficacy of MRNS in predicting the perioperative outcomes of OPN. Sixty-four cases were included in the study. Demographic parameters, tumor characteristics, and perioperative outcomes were analyzed. Correlation with the post-operative outcomes and the strengths of MRNS were compared with various other nephrometry scores.

Results

The mean age of the patients was 52.89 years, 60.9% were male and 53.1% had a right-sided mass. The comorbidities, body mass index, and performance scores were evenly distributed across the complexity groups (P > 0.05). The mean tumor size was 4.13 cm and the mean MRNS and RNS were 9.45 and 6.1, respectively. 60.9% of the cases had no complications. Major complications (Clavien-Dindo grade [CDG] 3+) were noted in five cases (7.8%). The trifecta of neargin, ischemia, and complications (MICs) score was achieved in 85.9% and was achieved in 71.9% of the cases. MRNS was found to be an independent predictor of the trifecta outcomes (P = 0.04). Receiver-operating characteristic curve of MRNS analyzing the major complications as per the CDG showed an area under the curve of. 804, indicating good prediction of complications by the MRNS.

Conclusions

MRNS improves the predicting power of RNS by attributing enhanced scores to key elements and by adding new elements. Also, MRNS has good ability to predict the achievement of the trifecta and MIC.

Trifecta and pentafecta outcomes following robot-assisted partial nephrectomy in a multi-institutional cohort of Indian patients

Introduction

The literature on studies reporting trifecta or pentafecta outcomes following robot-assisted partial nephrectomy (RAPN) in Indian patients is limited. The primary aim of this study was to report and evaluate the factors predicting trifecta and pentafecta outcomes following RAPN in Indian patients using the multicentric Vattikuti collective quality initiative (VCQI) database.

Methods

From the VCQI database for patients who underwent RAPN, data for Indian patients were extracted and analyzed for factors predicting the achievement of trifecta and pentafecta following RAPN. Trifecta was defined as the absence of complications, negative surgical margins, and warm ischemia period shorter than 25 min or zero ischemia. Pentafecta covers all the trifecta criteria as well as >90% preservation of estimated glomerular filtration rate (eGFR) and no stage upgrade of chronic kidney disease at 12 months.

Results

In this study, among 614 patients, the trifecta was achieved in 374 patients (60.9%) and pentafecta was achieved in 24.2% of the patients. Patients who achieved trifecta had significantly higher mean age (54.1 vs. 51.0 years, P = 0.005), body mass index (BMI) (26.7 vs. 26.03 kg/m2, P = 0.022), and smaller tumor size (38.6 vs. 41.4 mm, P = 0.028). The preoperative eGFR (84.2 vs. 91.9 ml/min, P = 0.012) and renal nephrometry score (RNS) (6.96 vs. 7.87, P ≤ 0.0001) were significantly lower in the trifecta group. Comparing patients who achieved pentafecta to those who did not, we noted a statistically significant difference between the two groups for tumor size (36.1 vs. 41.5 mm, P = 0.017) and RNS (6.6 vs. 7.7, P = 0.0001). On multivariate analysis, BMI and RNS were associated with trifecta outcomes. Similarly, only RNS was identified as an independent predictor of pentafecta.

Conclusions

RNS and BMI were independent predictors of the trifecta. At the same time, RNS was identified as an independent predictor of pentafecta following RAPN.

Nephrometry scoring systems: their importance for the planning of nephron-sparing surgery and the relationships among them

In recent years, the development of new imaging techniques and scoring systems have improved the diagnosis and management of small renal masses. Imaging-based nephrometry scoring systems play an interesting role in the planning of nephron-sparing surgery, providing surgeons with the information necessary to determine the complexity of the renal mass, to deliver the appropriate postoperative care, and to predict adverse outcomes. The aim of this study was to review nephrometry scoring systems, evaluating their characteristics and the relationships among them. The urology and radiology communities should decide which nephrometry scoring system will prevail and be used in daily practice.

Patient REnal and Tumor Attribute Score (PRETA Score): a comprehensive renal nephrometry score for use in patients with renal masses planned for minimally invasive nephron sparing surgery

Few comprehensive nephrometry scoring systems correlate renal morphology and patient factors like presence of chronic kidney disease (CKD) with the trifecta outcomes after partial nephrectomy (PN). We have proposed a new nephrometry score that correlates these factors to perioperative outcomes in minimally invasive PN. Records of all patients who underwent robot assisted laparoscopic PN at our institute were retrospectively reviewed. Baseline patient characteristics and imaging features were noted. Centrality Index or C-Index score and Mayo Adhesive Probability (MAP) score were calculated. Intraoperative and post-operative outcomes along with trifecta outcomes of PN were also noted. A total of 63 patients underwent PN, of which 50 had positive and 13 had negative trifecta outcomes. A total of 22 patients developed post-operative complications, with the majority being Clavien-Dindo Grade 1 and Grade 2 complications. Size of the tumor, presence of CKD, Centrality Index, Mayo Adhesive Probability score, nearness to hilum, and presence of congenital anomaly of the kidney or solitary kidney were significant predictors of negative trifecta outcomes on multivariate analysis. Based on these parameters we developed the Patient REnal Tumor Attribute (PRETA) score. The score ranges from 6 to 14. Patients with a score of 6 were at low risk of negative trifecta outcomes, those with a score of 7-10 were at moderate risk, and those with a score > 10 were at high risk. Prospective validation of this score was then done on the next 15 consecutive PN patients. In developing this score, we have done away with redundant parameters and have incorporated patient factors which have an important effect on patient outcomes after minimally invasive PN.

Trifecta and pentafecta outcomes following robot-assisted partial nephrectomy for hilar versus nonhilar tumors: A propensity-matched analysis

Introduction

Hilar tumors are a unique subset of complex renal masses posing a potential surgical challenge during partial nephrectomy. The outcomes of hilar masses have not been compared to non-hilar renal masses of similar RENAL nephrometry score (RNS). In this study, we analyzed the outcomes of hilar versus nonhilar masses after a propensity score matching.

Methods

Prospectively maintained database of patients who underwent robot assisted PN between November 2014 and December 2018 was abstracted for hilar and nonhilar tumors. We performed propensity matching for baseline variables such as age, sex, body mass index, comorbidities, preoperative glomerular filtration rate, and RNS for each patient on the basis of propensity scores.

Results

We included 48 patients with hilar tumors and 153 with nonhilar tumors. On propensity matching, 41 patients were included in each group. The mean operative time (162.4 ± 48.9 min vs. 144.1 ± 38.8 min, P = 0.48), warm ischemia time (29.0 ± 8.8 min vs. 24.4 ± 8.2 min, P = 0.12), and the estimated blood loss (201.8 ± 184.7 ml vs. 150.6 ± 160.5 ml, P = 0.37) were not significantly different between the hilar and the nonhilar groups. Trifecta was achieved in only 14/41 (34.1%) of the patients in the hilar group as compared to 24/41 (58.5%) in the nonhilar group (P = 0.027). Logistic regression analysis identified that hilar location of the tumors was not an independent predictor of overall complications (OR 6.37, confidence interval [CI] 0.5-69.4, P = 0.4), trifecta (OR 0.38, CI 0.14-1.0, P = 0.051), and pentafecta outcomes (OR 0.4, CI 0.1-1.51, P = 0.17).

Conclusions

Hilar location was associated with poorer trifecta outcomes compared to the nonhilar tumors. However, hilar location per se was not an independent predictor of overall complications and trifecta and pentafecta outcomes.

Predicting Strict Trifecta Outcomes after Robot-Assisted Partial Nephrectomy: Comparison of RENAL, PADUA, and C-Index Scores

Nephrometry scores are designed to characterize tumors and stratify the surgical complexity. It remains unclear as to which nephrometry score can accurately predict the surgical outcomes. We aimed to assess the utility of radius, exophytic/endophytic, nearness, anterior/posterior, location (RENAL), preoperative aspects and dimensions used for anatomic classifications (PADUA), and centrality index (C-index) nephrometry scores for predicting the strict Trifecta achievement from a single institution series robotic-assisted partial nephrectomy (RAPN). We retrospectively identified the prospectively maintained robotic surgery database records of 91 patients who underwent RAPN between June 2015 and September 2020 in Antalya Training and Research Hospital. The main outcome of the study was the achievement of strict Trifecta (negative surgical margin, no major urologic complications, warm ischemia time ≤25 min, and ≥85% preservation of estimated glomerular filtration rate). A multivariable analysis was performed to identify the factors of strict Trifecta success. The mean patient age was 55.82 ± 13.37 years with a median clinical tumor size of 3.5 cm (IQR 2.5–4.9). The median RENAL, PADUA, and C-index score were 7(IQR 6–8), 8(IQR 7–10), and 2.01(IQR 1.64–2.72), respectively. A strict Trifecta could be achieved in 54 patients (59.3%). Clinical tumor size (P = 0.011), RENAL risk groups (low:reference; intermediate; P = 0.040; high; P = 0.009), PADUA risk groups (low:reference; intermediate; P = 0.044; high; P = 0.001) and C-index risk groups (low:reference; high; P = 0.015) were the independent predictors of strict Trifecta attainment in the multivariate analysis. None of the nephrometry scores were a superior predictor compared to other nephrometry scores in comparative analysis. RENAL, PADUA, and C-index scores were all independent predictors of a strict Trifecta achievement. Our comprehensive comparison of the three scores identified that none of the nephrometry scores proved to be inferior to others nephrometry scores.

A comparison of DAP and RENAL scoring systems in the achievement of trifecta in laparoscopic partial nephrectomy

PURPOSE

The concept of 'trifecta' outcome postulated for radical prostatectomy has been adopted for partial nephrectomy, the gold standard for management of small renal masses. We sought to evaluate the role of nephrometry scores in predicting outcomes in terms of the trifecta. We compared two scoring systems for renal tumour complexity (RENAL and DAP) in the prediction of trifecta outcomes.

MATERIALS AND METHODS

Sixty-nine patients who underwent laparoscopic PN (LPN) were evaluated in a single-centre retrospective study (2010-2017). RENAL and DAP scores were measured. Parameters relevant to the trifecta were tabulated.

RESULTS

When comparing the two scoring systems in terms of warm ischaemia time (WIT), the DAP score could predict with statistically significant accuracy the completion of resection within 25 mins of WIT. Tumours were more evenly distributed according to anatomical characteristics with the DAP scoring system than with the RENAL scoring system. When comparing these systems in terms of complications, neither predicted complications based on complexity with significant accuracy. A low RENAL score predicted trifecta achievement in three-fourth (71.4%) of patients, while a medium RENAL score predicted trifecta achievement in half (54%) of patients. DAP score predicted trifecta achievement in all tumours with a low score, two-third (66%) in medium and less than half (42%) with a high score. Predictions based on DAP were accurate and significantly so (p = 0.024).

CONCLUSIONS

DAP score predicted the outcomes of LPN in terms of trifecta significantly better than the RENAL score. In our experience, the DAP score was able to distribute tumour complexity among its groups more effectively than the RENAL score. There is early evidence that the DAP score may be more useful than the RENAL score for decision-making in nephron sparing surgery. This is especially pertinent for small renal masses at the upper limits of tumour complexity for which minimally invasive techniques can be safely applied.

A Computed Tomography Angiography (CTA): Based score (SGPGI score) to predict level of difficulty, operative, and postoperative outcomes for right-sided radical nephrectomy

OBJECTIVE

With the advent of laparoscopic approach for the large (T1b-T3a ± N1) right renal masses, higher rates of complications and conversion to open surgery are being reported. The role of preoperative angioembolization (PAE), which has increased cost and inherent morbidity but may help in select circumstances has also not been clearly defined in the literature. We therefore devised a scoring system (SGPGI score) based on pre-operative Computed Tomography Angiography (CTA) to predict the level of difficulty of radical nephrectomy and enhance its safety and efficacy which could also be used for the judicious use of PAE in selected cases.

METHODS

In a prospective observational study on 52 patients with right renal masses from January 2014 to July 2018, we calculated a score based on CTA parameters. The patients were stratified for type and duration of surgery, blood loss, postoperative stay, and Clavien-Dindo grade of postoperative complications.

RESULTS

Patients were classified into three groups based on our scoring system. Progressively groups with higher score had higher blood loss, operating time, complications and hospital stay, and were more likely to have undergone conversion to open surgery (Area under curve 0.8625 for a cut off score of 10). Intraclass Correlation Coefficient (ICC) was 0.678-1 for the different components of our score.

CONCLUSION

The pre-operative CTA based SGPGI score evaluates right renal masses and is able to predict intra-operative difficulties effectively, leading to enhancement of surgery safety and efficacy. It also helps judiciously use PAE.

Parallel comparison of R.E.N.A.L., PADUA, and C-index scoring systems in predicting outcomes after partial nephrectomy: A systematic review and meta-analysis

Objective

To parallelly compare the applicability of the radius, exophytic/endophytic, nearness, anterior/posterior, location nephrometry score (R.E.N.A.L.), the Preoperative Aspects and Dimensions Used for an Anatomical (PADUA), and the centrality index (C‐index) scoring systems in predicting clinical outcomes after partial nephrectomy (PN).

Methods

We searched EMBASE, PubMed, Ovid, and Web of Science to perform a meta‐analysis examining the correlation coefficients between three nephrometry scores (NSs) and warm ischemia time (WIT), estimated blood loss (EBL), operation time (OT), length of stay (LOS), and absolute change in eGFR (ACE) up to 25 January 2021.

Results

In total, 13 studies including 1496 patients met the criteria for further analysis. Overall, all scoring systems had statistically significant correlations with the WIT, EBL, OT, ACE and LOS and ACE, except for the correlation between PADUA and LOS (r = 0.16 [−0.00, 0.31], p > 0.05). The C‐index had the strongest correlation with WIT (r = −0.35 [−0.43, −0.26], p < 0.05) and ACE (r = −0.29 [−0.48, −0.10], p < 0.05). Weak correlations were observed between OT as well as EBL and each scoring system. Publication bias was observed in PADUA score predicting ACE (p = 0.04) and high heterogeneity was found in some of our results.

Conclusion

Until now, this is the first meta‐analysis that parallelly compares these three scoring systems in predicting outcomes after PN. We found that all NSs showed a statistically significant correlation with WIT, EBL, OT, and ACE. Moreover, the C‐index scoring system is the best predictor of WIT and ACE. Due to the existence of publication bias and high heterogeneity, more well‐designed and large‐scale studies are warranted for validation.

To our knowledge, this is the first meta‐analysis that parallelly compares these three scoring systems in predicting outcomes after PN. Overall, three scoring systems were significantly correlated with WIT, EBL, OT and ACE. Moreover, the C‐index scoring system outperformed R.E.N.A.L. and PADUA scoring systems in WIT and ACE.

External validation of SPARE nephrometery score in predicting overall complications, trifecta and pentafecta outcomes following robot-assisted partial nephrectomy

BACKGROUND

There is an ongoing need and search for a simple yet accurate nephrometry scoring system for predicting the postoperative outcomes after partial nephrectomy (PN). Simplified PADUA Renal (SPARE) Nephrometry Scoring System, a simplified version of Preoperative Aspects and Dimensions Used for an Anatomical Classification (PADUA) has been proposed as a predictor of postoperative complications following PN recently. However, this score has never been externally validated and assessed as a predictor of trifecta and pentafecta outcomes of PN. In the current study, we applied the SPARE scoring system to our robot-assisted PN cohort (RAPN).

METHODS

Prospectively maintained data of patients, who underwent RAPN from November 2014 to December 2018, was abstracted. Imaging was analyzed to calculate SPARE and RENAL nephrometry scores (RNS) by two urologists, independently. SPARE was compared with complications, trifecta outcomes, pentafecta outcomes, and RENAL nephrometry scoring (RNS).

RESULTS

Data of 201 RAPN patients were analyzed. The mean SPARE score was 3 (range 0-11). One hundred thirteen patients were classified as low risk, 64 as intermediate risk, and 24 as high risks. On multivariate analysis SPARE score alone predicted complications (OR=1.37, P=0.014) and trifecta outcomes (OR=0.75, P=0.000) while age (OR=0.96, P=0.042), preoperative eGFR (OR=0.97, P=0.001) and SPARE scores (OR=0.81, P=0.016) were predictors for pentafecta outcomes. Receiver operated curve (ROC) analysis between SPARE and RNS in predicting the complications; trifecta and pentafecta outcomes had a comparable area under the curve.

CONCLUSIONS

Our study validates the SPARE nephrometry scoring system in predicting postoperative complications, trifecta, and pentafecta outcomes in a RAPN cohort. The predictive accuracy of SPARE is similar to RNS.

The associations of RENAL, PADUA and C-index nephrometry scores with perioperative outcomes and postoperative renal function in minimally invasive partial nephrectomy

OBJECTIVE

This study aimed to assess the utility of the radius, exophytic/endophytic, nearness, anterior/posterior, location (RENAL); preoperative aspects and dimensions used for an anatomic evaluation (PADUA), and centrality index (C-index) scores for the outcomes of partial nephrectomy (PN).

MATERIAL AND METHODS

The patients who underwent PN with contrast-enhanced preoperative imaging from January 2015 to June 2018 were identified. The RENAL, PADUA, and C-index scores were assigned. The correlation between these scoring systems and perioperative and long-term renal functional outcomes were evaluated.

RESULTS

A total of 78 patients were included in the study (58 men and 20 women; age, 58±11.4 years). Median warm ischemia time (WIT), estimated blood loss (EBL), and operation time (OT) were 26 min, 115 mL, and 140 min, respectively. The RENAL score was related to WIT, EBL, and OT (p<0.001, p=0.003, and p=0.023, respectively). The PADUA score was associated with WIT, EBL, and OT (p<0.001, p=0.013, and p=0.005, respectively). The C-index score was correlated with WIT, EBL, and OT (p<0.001, p=0.010, and p=0.001, respectively). The C-index score also correlated with the percentage change in the estimated glomerular filtration rate (p=0.037). However, on univariable and multivariable regression analyses, only WIT significantly affected the postoperative estimated glomerular filtration rate reduction.

CONCLUSION

The RENAL, PADUA, and C-index scores were significantly associated with perioperative outcomes of PN. In addition, the C-index score was correlated with long-term renal functional outcomes.