Residual kidney function (RKF) is an important prognostic indicator in peritoneal dialysis (PD) patients. So far, there are no prediction tools available for RKF, and the association between serum bicarbonate and RKF has received little attention in patients with PD. We aimed to develop a nomogram for the preservation of RKF based on the time-averaged serum bicarbonate (TA-Bic) levels. A prediction model was established by conducting a retrospective cohort study of 151 PD patients who had been treated at the First Affiliated Hospital of Anhui Medical University. The nomogram was developed using a multivariate Cox regression model. The discrimination, calibration, and clinical utility of the model were evaluated by the C-index, receiver operating curve (ROC) curve, calibration curve, and decision curve analysis. In the elderly PD onset, higher baseline values of residual glomerular filtration rate, total Kt/V and higher TA-Bic levels were identified as protective predictors of RKF loss. The nomogram was conducted on the basis of the minimum value of the Akaike Information Criterion and Bayesian Information Criterion with a reasonable C-index of 0.766, showing great discrimination, proper calibration, and high potential for clinical practice. Through the total score of the nomogram, the patients were classified into the high-risk group and low-risk group, and a higher cumulative incidence of complete RRF loss was found in the high-risk group compared with the patients in the low-risk group. The novel predictive nomogram model can predict the probability of RKF preservation in long-term PD patients with high accuracy. Future studies are needed to externally validate the current nomogram before clinical application.