The value of nomogram model combined with contrast-enhanced ultrasound in the differential diagnosis of cervical tuberculosis lymphadenitis and metastatic lymph node.

Background: This study aimed to construct an effective Nomogram for the differential diagnosis of cervical tuberculosis lymphadenitis (CTBL) and cervical metastatic lymph node (CMLN) based on ultrasound (US).

Methods: Retrospectively analyzed 227 patients with CTBL and CMLN who attended Hangzhou Red Cross Hospital from January 2020 to October 2022, and were divided into a training set (n = 163) and a validation set (n = 64) according to the clinical data, US, and CEUS qualitative and quantitative analysis data were recorded to establish the prediction model and perform validation. The area under curve (AUC) of the receiver operating characteristic curve (ROC) was used to assess the discrimination of the model; the calibration curve and brier coefficient were used to assess the calibration of the model; and a Nomogram prediction model was constructed to visualize the results nomogram prediction model was constructed to visualize the results.

Results: Gender (OR = 0.200, 95% CI:0.090-0.470, P < 0.001), age (OR = 0.170, 95% CI:0.070-0.410, P < 0.001), liquefaction necrosis (OR = 2.560, 95% CI:1.080-6.040, P = 0.033), perfusion defect (OR = 2.570, 95% CI:1.010-6.580, P = 0.048), and standard deviation (StdDev) (OR = 3.040, 95% CI:1.220-7.570, P = 0.017) were the independent predictors of the constructed model. The AUCs of the constructed predictive model in the training set and validation set were 0.844 and 0.927, respectively; from the calibration curves, it was observed that the predicted values of the model and the actual observed values fell near the 45° diagonal, and the brier scores were 0.145 and 0.109 in the training set and validation set, respectively.

Conclusion: StdDev combined with gender, age, and the presence of liquefaction necrosis and perfusion defects are important features to identify CTBL and CMLN, and the constructed visual nomogram is intuitive and convenient to improve the efficiency of clinical work.