Objective To investigate the influencing factors of delayed extubation after arthroplasty under general anesthesia and develop a nomogram model. Methods A retrospective analysis was performed of the electronic medical records of 377 elderly patients with hip or knee arthroplasty under general anesthesia from January 2016 to December 2020. According to whether the extubation time of tracheal catheter exceeded 1 h, the patients were divided into normal group (n=297 ) and delayed group (n=80). R Vision 3.6.1 and SPSS statistics 25.0 were used for data analysis. Depending on data type, t test, Mann-Whitney U test, χ² test or Fisher′s exact probability method were used for comparison between groups. Multivariate Lasso regression analysis was used to screen the risk factors of postoperative delayed extubation, and a nomogram was developed and validated. Results Multivariate Lasso regression analysis showed that being aged, hyperlactacidemia, and coefficient of variation (CV) were positively correlated with delayed extubation in arthroplasty (OR>1; P<0.05), and correction of hypokalemia and intraoperative blood glucose control (final blood glucose ＜8.3 mmol/L) were negatively correlated with delayed extubation (OR<1; P<0.05). A nomogram was developed, and Harrel′s C-index [the area under the receiver operating characteristic curve (AUC), 0.800,95%CI 0.745-0.854) and Hosmer-Lemeshow deviation test (χ²=3.000; P=0.600) confirmed good prediction efficiency of the model, with stable internally validated AUC value and calibration curve. The decision curve of the model showed significant net benefit and good clinical practicability. Conclusion Being aged, hyperlactacidemia, and CV were independent risk factors for delayed extubation after arthroplasty, while correction of hypokalemia and blood glucose control were protective factors. The nomogram is of high clinic reference value for delayed extubation.