The pathological pathways involved in the cardiorenal syndrome remains unknown and more research is needed to uncover this relationship. Here, we investigate the correlation of electrocardiogram (ECG) features in chronic heart failure (CHF) patients with renal function measured by estimated glomerular filtration rate (eGFR) We used the Sudden Cardiac Death in Chronic Heart Failure (MUSIC) data from Physionet database, to classify CHF patients into two classes (eGFR < 66 and eGFR > 66). Different machine learning (ML) models were employed, among which, the logistic regression (LR) model demonstrated the best performance, achieving an accuracy of 69.35%, a sensitivity of 70.71%, and an area under the receiver operating characteristic curve (ROC-AUC) of 0.752. Furthermore, explainable ML analysis revealed that Minimum RR Interval (RRI), QRS Duration, RR Range, and corrected (QTc) were identified as significant predictors of eGFR classes. These results pave the way for understanding the cardiorenal syndrome.