Nomogram to predict postoperative cognitive dysfunction in elderly patients undergoing gastrointestinal tumor resection.

OBJECTIVE

To establish a nomogram model for the prediction of postoperative cognitive dysfunction (POCD) in elderly patients undergoing gastrointestinal tumor resection.

METHODS

A total of 369 elderly patients scheduled for elective gastrointestinal tumor resection under general anesthesia were included. The cognitive function of each participant was assessed by the Mini-Mental State Examination (MMSE) 1 day before surgery and 7 days after surgery for the diagnosis of POCD. According to the results, patients were divided into a POCD group and a non-POCD group. The differences in hospitalization data and examination results between the two groups were compared. A logistic regression model was used to explore the risk factors for POCD in elderly patients undergoing gastrointestinal tumor resection, and a nomogram was then constructed based on these factors. The diagnostic performance of the nomogram was evaluated using the area under the receiver operating characteristic curve (AUROC) and a calibration plot. The clinical usefulness of the nomogram was estimated using decision curve analysis (DCA).

RESULTS

Among the 369 patients undergoing gastrointestinal tumor resection, 79 patients had POCD, with a positive rate of 21.4%. The nomogram model comprised the following variables: age, body mass index (BMI), history of cerebrovascular disease, preoperative white blood cell (WBC) count, preoperative hemoglobin (Hb) level, intra-operative blood loss, and operation time. The model showed good discrimination, with an area under the curve (AUC) of 0.710 (95% CI = 0.645-0.775), and good calibration (Hosmer-Lemeshow test, χ = 5.133, = 0.274). Internal validation also maintained ideal discrimination and calibration. Decision curves indicated that when the threshold probability was above 0.1, the nomogram achieved more benefit than both the treat-all and treat-none policies.

CONCLUSION

This scoring system is the first nomogram model developed for the prediction of POCD in elderly patients undergoing gastrointestinal tumor resection. It has good efficacy in the prediction of POCD risk and could provide an important reference for the prevention, management, and treatment of POCD.