Objective To explore the energy spectrum CT parameters, clinicopathological factors, influencing factors and the application efficiency of the prediction model of lymph node metastasis in patients with non-small cell lung cancer (NSCLC) at high altitude area.

Methods A total of 84 NSCLC patients from January 2020 to July 2021 in Qinghai Provincial People′s Hospital were selected as research objects, of whom 50 patients were conducted with energy spectrum CT detection and their clinical materials were collected. According to the pathological examination results of lymph node metastasis, 50 patients were divided into metastasis group and non-metastasis group, the differences of clinicopathological factors and energy spectrum CT parameters between the two groups were analyzed by univariate analysis, the influencing factors of lymph node metastasis in NSCLC patients were analyzed by multivariate regression analysis, and a prediction model was established. The prediction model was used to analyze the lymph node metastasis of the other 34 NSCLC patients, and the clinical test results were used as the gold standard to evaluate the application value of the prediction model.

Results ① The incidence of lymph node metastasis in 50 patients was 30.00%. The preoperative carcinoembryonic antigen (CEA), ratio of central carina position and lymph node diameter in the metastasis group were significantly higher than those in the non-metastasis group (P < 0.05). ② Slope of energy spectrum curve of lymph nodes (λHu), normalized iodine concentration (NIC) and λHu ratio of lymph nodes to primary lesions in the metastasis group were significantly lower than those in the non-metastasis group (P < 0.05). ③ The risk factors of lymph node metastasis in NSCLC patients were the λHu ratio of lymph nodes to primary lesions, preoperative CEA≥5 ng/mL and lymph node diameter≥3 cm. ④ The accuracy, specificity and sensitivity of the prediction model for lymph node metastasis in 34 patients with NSCLC were all above 80.00%.

Conclusion In the NSCLC patients with lymph node metastasis, energy spectrum CT parameters and related risk factors in clinicopathological factors can be used to establish a prediction model for accurate prediction, and this model can be used as the preferred tool to predict lymph node metastasis.