Serum Procalcitonin, Smoking History Combined Age Established a New Prediction Model for Predicting Dynamic Changes of Chest CT Images in Adult Community-Acquired Pneumonia (CAP) Patients.

Abstract

Chest CT is widely used in clinical diagnosis and efficacy evaluation of CAP. While repeated chest CT examinations to evaluate dynamic changes in chest CT images in a short period of time is a common phenomenon, it causes a lot of waste of medical resources, and due to the large dose of CT radiation, it can cause some harm to the human body. The purpose of this study is to establish a new model to predict the dynamic chest CT image changes of CAP patients by analyzing the age, smoking history, and serum inflammatory markers. This is a retrospective study. All patients had received chest CT scan and serum inflammatory indexes were measured, including procalcitonin (PCT), high-sensitivity C-reactive protein (hs-CRP), white blood cell (WBC) and erythrocyte sedimentation rate (ESR). The second chest CT examination was performed after a week of treatment. General information on the medical record was also recorded (including age, smoking history, drinking history, and others). Main outcome measures were the changes of chest CT images, including absorption and non-absorption (including patients with progressive inflammation). Single factor analysis and two-dimensional logistic regression analysis were used to explore the independent risk factors of the new CT image change prediction model for CAP patients. ROC was used to evaluate the sensitivity and specificity of the new model. Among 220 patients with CAP, 150 patients had absorption in chest CT after a week of treatment (150/220), the remaining 70 patients had no absorption or even progression (70/220). Age, PCT, and smoking history were independent risk factors for inflammatory absorption. The AUC of ROC curve was 0.89 (95% CI 0.83 - 0.94), the sensitivity was 88.70%, and the specificity was 80.00%. A new prediction model consists of serum PCT, age, and smoking history has high specificity and sensitivity in predicting dynamic CT changes in adult CAP patients.