Search

Background: One of the most prevalent complications of Partial Nephrectomy (PN) is Acute Kidney Injury (AKI), which could have a negative impact on subsequent renal function and occurs in up to 24.3% of patients undergoing PN. The aim of this study was to predict the occurrence of AKI following PN using preoperative parameters by applying machine learning algorithms., Methods: We included all adult patients (n = 723) who underwent open PN in our department since 1995 and on whom we have data on the pre-operative renal function. We developed a random forest (RF) model with Boolean satisfaction-based pruned decision trees for binary classification (AKI or non-AKI). Hyper-parameter grid search was performed to optimize the model's performance. Fivefold cross-validation was applied to evaluate the model. We implemented a RF model with greedy feature selection to binary classify AKI and non-AKI cases based on pre-operative data., Results: The best model obtained a 0.69 precision and 0.69 recall in classifying the AKI and non-AKI groups on average (k = 5). In addition, the model's probability to correctly classify a new prediction is 0.75. The proposed model is available as an online calculator., Conclusions: Our model predicts the occurrence of AKI following open PN with (75%) accuracy. We plan to externally validate this model and modify it to minimally-invasive PN., (© 2022. The Author(s).)

Background: This study aims to prospectively evaluate the ability of Neutrophil-to-Lymphocyte ratio (NLR) to forecast recurrence in patients with non-muscle invasive bladder cancer (NMIBC). This is a continuation of our two previous retrospective studies that indicated the NLR > 2.5 criterion as a predictor of recurrence in patients with NMIBC., Methods: Since December 2013, all patients admitted to our department for TUR-BT and agreed to participate, had a blood drawn for cell count and differential 24 h prior to surgery. Patients with pathological NMIBC were followed prospectively for disease recurrence. The end-point of the follow up was either a cancer recurrence or the termination of the study. Univariate and multivariate Cox regressions were performed to assess the NLR > 2.5 predictive capability for recurrence, versus and in conjunction to the pathologically based EORTC score, among additional statistical analyses., Results: The study cohort included 96 men and 17 women with a median age of 72 years. Sixty-four patients (56.6%) have had a recurrence during the study occurring at the median time of 9 months (IQR 6, 13), while the median follow-up time for patients without recurrence was 18 months (IQR 10, 29). Univariate Cox regressions for recurrence demonstrated significance for NLR > 2.5 for the whole cohort (p = 0.011, HR 2.015, CI 1.175-3.454) and for the BCG sub-group (p = 0.023, HR 3.7, CI 1.2-11.9), while the EORTC score demonstrated significance for the 'No Treatment' subgroup (p = 0.024, HR 1.278, CI 1.03-1.58). When analyzed together as a multivariate Cox model, the NLR > 2.5 and EORTC score retained their significance for the aforementioned groups, while also improving the EORTC score significance for the whole cohort., Conclusion: NLR > 2.5 was found to be a significant predictor of disease recurrence and demonstrated high hazard ratio and worse recurrence-free survival in patients with NMIBC, especially in those treated with BCG. Additionally, our data demonstrated statistical evidence that NLR > 2.5 might have an improving effect on the EORTC score's prediction when analyzed together.