Anxiety in patients with trigeminal neuralgia and the predictive modeling
Objective To investigate anxiety in patients with trigeminal neuralgia and to develop a predictive model for it. Method 302 patients with trigeminal neuralgia visited our hospital from February 2019 to December 2023 were selected for the study. The patients were divided into 211 cases in the model group and 91 cases in the validation group in a 7:3 ratio.Basic condition,sleep quality,social support,coping styles and other relevant factors that may affect the anxiety of patients with trigeminal neuralgia were collected. Additionally,patients in the model group were divided into anxious and non-anxious groups according to the presence or absence of anxiety. The indexes of the patients in the 2 groups were compared,and multifactorial logistic regression was carried out after the identification of potentially affecting factors using LASSO regression,and the columnar plot model was set up and verified. Result Of the 211 patients with trigeminal neuralgia in the model group,63 (29.86％) developed anxiety. The anxiety group had a greater age,proportion of monthly household income<3000 RMB,proportion of unmarried status,Visual Analogue Scale (VAS) score and Pittsburgh Sleep Quality Index (PSQI) score than the non-anxiety group. In the anxiety group,the proportions of junior high school education and lower,single painful nerve branches,health insurance payment,disease duration<1 year,each pain duration<1 min,episodes with a frequency of<10 episodes/d,social support rating scale (SSRS) scores,and positive coping styles were smaller than those of the non-anxiety group (P<0.05).The results of multifactorial logistic regression analysis performed on the basis of LASSO regression showed that age,painful nerve branches,disease duration,VAS,PSQI,and SSRS were independent factors of anxiety status in patients with trigeminal neuralgia (P<0.05). The results of model validation showed the following:The model group's performance was evaluated with an area under the ROC curve of 0.832,which had a 95％ confidence interval (CI) of 0.766 to 0.897. It showed a sensitivity of 84.1％ and a specificity of 75.0％. In comparison,the validation group had an area under the ROC curve of 0.786,with a 95％ CI ranging from 0.716 to 0.855,a sensitivity of 77.8％,and a specificity of 68.9％. Calibration curves for both the model and validation groups had slopes of 1 and intercepts of 0.000,indicating that the model curves were essentially fitted to the ideal model on a diagonal. The clinical validity analysis showed that the model's net benefit for predicting anxiety in patients with trigeminal neuralgia was highest when the threshold for the predictive probability was set between 0.15 and 0.95. Conclusion Anxiety in patients with trigeminal neuralgia is mainly affected by increasing age,increased number of painful nerve branches,prolonged duration of the disease,higher VAS scores,higher PSQI scores,and lower SSRS scores. The column-line graphical model developed in this study can be used to predict the risk of anxiety in patients with trigeminal neuralgia.
Trigeminal neuralgiaAnxietyMultifactorial analysisColumnar graphical model
NETL
NSTL
万方数据
