目的·基于超声评分参数及临床指标，构建鉴别原发性干燥综合征（primary Sjὅgren′s syndrome，PSS）与免疫球蛋白G4相关唾液腺炎（immunoglobulin G4-related sialadenitis，IgG4-RS）列线图并评估其性能。方法·回顾性选择2018年1月—2023年12月就诊于上海交通大学医学院附属第九人民医院的PSS患者141例和IgG4-RS患者31例。收集患者的超声评分参数，包括腮腺超声（parotid gland ultrasound，PGUS）评分、下颌下腺超声（submandibular gland ultrasound，SMGUS）评分、唾液腺超声（salivary gland ultrasound，SGUS）评分，及临床指标包括性别、年龄、抗SSB/La抗体、抗SSA/Ro60抗体、抗SSA/Ro52抗体、IgG、类风湿因子（rheumatoid factor，RF）。利用最小绝对收缩和选择算子（least absolute shrinkage and selection operator，LASSO）回归筛选出最优超声评分参数和临床指标，建立PSS和IgG4-RS的鉴别诊断列线图。通过bootstrap法进行模型的内部验证。分别采用受试者操作特征（receiver operator characteristic，ROC）曲线、校准曲线及决策曲线分析（decision curve analysis，DCA）评价模型的区分度、校准度及其在临床中的应用价值。结果·通过LASSO回归算法，共筛选出性别、年龄、抗SSA/Ro60抗体、抗SSA/Ro52抗体、PGUS评分、SMGUS评分6个变量，根据该6个指标建立列线图模型。该列线图模型的ROC曲线显示曲线下面积（area under the curve，AUC）为0.976，具有较强的区分度。bootstrap法内部重复抽样1 000次进行验证，平均绝对误差0.018，校准曲线表明预测值与实测值良好吻合。DCA表明该列线图具有一定的临床实用性。结论·基于超声评分参数及临床指标建立的列线图，在鉴别PSS和IgG4-RS方面展示了良好的区分度和校准度，有望为临床鉴别诊断2种疾病及制定相应治疗方案提供参考。

Objective ·To construct and validate a nomogram for distinguishing primary Sjὅgren′s syndrome (PSS) from immunoglobulin G4-related sialadenitis (IgG4-RS) based on ultrasound scoring parameters and clinical indicators. Methods ·A total of 141 patients with PSS and 31 patients with IgG4-RS were retrospectively recruited from Shanghai Ninth People′s Hospital, Shanghai Jiao Tong University School of Medicine from January 2018 to December 2023. The ultrasound scoring parameters, including parotid gland ultrasound (PGUS) score, submandibular gland ultrasound (SMGUS) score, and salivary gland ultrasound (SGUS) score, along with clinical indicators such as gender, age, anti-SSB/La antibody, anti-SSA/Ro60 antibody, anti-SSA/Ro52 antibody, IgG, and rheumatoid factor (RF), were collected. The optimal US scoring parameters and clinical indicators were screened by least absolute shrinkage and selection operator (LASSO) regression, and a nomogram model was constructed to distinguish PSS from IgG4-RS. The internal validation of the model was carried out through bootstrap method. Receiver operator characteristic (ROC) curve, calibration curve and decision curve analysis (DCA) were used to estimate the discrimination, calibration, and clinical utility of the nomogram model, respectively. Results ·LASSO regression identified six major variables: gender, age, anti-SSA/Ro60 antibody, anti-SSA/Ro52 antibody, PGUS score, and SMGUS score. These variables were used to construct the nomogram. ROC curve of the nomogram showed that the area under the curve (AUC) was 0.976, indicating the nomogram had strong discrimination ability. The bootstrap method was used for internal validation with 1 000 resampling iterations, and the average absolute error was 0.018. Calibration curve demonstrated good agreement between predicted and observed values. DCA indicated that the nomogram had certain clinical utility. Conclusion ·The nomogram based on ultrasound scoring parameters and clinical indicators demonstrates excellent discrimination and calibration in differentiating PSS from IgG4-RS. It has the potential to assist in clinical diagnosis and decision-making.