Objective To explore the use of a novel inflammatory composite scoring system with a column chart model to predict the risk of postoperative delirium in patients with lumbar spine fractures. Methods 251 patients with lumbar spine fracture who had been admitted to our hospital and had undergone surgical treatment from June 2020 to December 2022 were selected as study subjects， and then divided into a postoperative delirium group （n = 55） and a delirium-free group （n = 196） according to the patients' conditions. The clinical data of the two groups were recorded and subjected to univariate analysis. ROC curve was used to analyze the predictive value of the indexes for postoperative delirium in the patients； logistic regression was used to analyse the independent risk factors for postoperative delirium and construct a columnar graph model； the calibration curves were used for the internal validation of the model； the consistency index （C-index） was calculated； and the decision curve was used to evaluate the predictive efficiency of the model. Results The differences between the two groups in terms of age， surgical duration， intraoperative bleeding， and inflammatory response markers NLR， PLR， and CAR were statistically significant （P < 0.05）. The AUCs of age， surgical time， intraoperative bleeding， NLR， PLR， and CAR were 0.788， 0.714， 0.797， 0.824， 0.839， and 0.811， respectively； and the AUC of the constructed N.P.Cs scoring scale was 0.963. Age， intraoperative bleeding， and N.P.Cs were independent risk factors affecting postoperative delirium in patients with lumbar spine fractures. The calibration curve of the column-line diagram prediction model was close to the original and ideal curves， with a C-index of 0.847 （95% CI：0.811 ~ 0.903）； the threshold value of the column-line diagram prediction model was greater than 0.17， providing a net clinical benefit. Conclusion Age， intraoperative bleeding， and N.P.Cs are independent risk factors for postoperative delirium in patients with lumbar spine fractures. The column chart prediction model based on this construction can provide guidance for the prevention and treatment of postoperative delirium in patients with lumbar spine fractures.