目的：分析CT眼眶结构定量分析的眼外肌厚度与甲状腺相关性眼病(TAO)的临床活动性评分(CAS)相关性。

方法：本研究为回顾性分析，选取2020年10月至2025年2月我院收治的TAO患者的临床资料。另选我院体检健康者作为对照组。所有参与者均接受CT检查，比较两组参与者眼上直肌、下直肌、内直肌、外直肌、眼眶面积、突眼度以及眼外肌总横截面积/眼眶面积比值(OM/TOA)情况，采用临床活动性评分(CAS)对TAO患者进行评估，分析CAS评分与CT眼眶结构定量分析指标的相关性。不同活动期 TAO患者CT眼眶结构定量分析指标比较，CT眼眶结构定量分析指标对活动期TAO患者的预测价值。

结果：本研究共纳入TAO患者77例，其中男38例、女39例； 年龄28-70(平均49.5±6.9)岁。对照组77例，其中男40例、女37例； 年龄26-70(平均49.0±7.3)岁。两组参与者性别、年龄比较均无差异(均P>0.05)。TAO组患者左眼、右眼及双眼CT眼眶结构定量分析各指标均明显高于对照组(均P<0.001)，TAO组患者CAS评分为3.94±1.51分； CAS评分与CT眼眶结构定量分析各指标均呈正相关(均P<0.05)。根据CAS评分结果，TAO患者CAS评分<3分为非活动期共14例28眼，其中男8例，女6例，平均年龄43.79±9.58岁。CAS评分≥3分为活动期共63例126眼，其中男30例、女33例，平均年龄50.78±5.47岁，不同活动期 TAO患者性别比较无差异(P=0.519)，年龄比较有差异(P<0.001)。非活动期患者CT眼眶结构定量各指标均明显低于活动期患者(P<0.05)。最终选取上直肌、年龄、突眼度纳入Logistic回归模型分析结果显示，上直肌、突眼度与TAO活动期存在相关性(P<0.05)，年龄与TAO活动期无相关性(P>0.05)。ROC曲线分析结果显示：曲线下面积(AUC)为0.863，标准误差为0.063，P<0.001，AUC的95%置信区间(95%CI)为0.740-0.985； 模型预测敏感度为73.0%，特异度为92.9%，约登指数为0.659； 预测精度为97.9%，召回率为73.0%，F1值为0.836； 预测最佳临界值为0.857，预测概率0.74。

结论：TAO患者可采用CT眼眶结构定量分析评估病情程度。

METHODS:This was a retrospective analysis, selecting clinical data of TAO patients admitted to the hospital from October 2020 to February 2025. Healthy individuals were chosen from hospital's physical examination as the control group. All participants underwent CT examination, the superior rectus muscle, inferior rectus muscle, medial rectus muscle, lateral rectus muscle, orbital area, protrusion degree, and total cross-sectional area of extraocular muscles/total orbital area ratio(OM/TOA)from the two groups of participants were compared. CAS was used to evaluate TAO patients, and the correlation between CAS score and quantitative analysis indicators of CT orbital structure was analyzed. Quantitative analysis indicators for CT orbital structure in TAO patients at different stages of activity were compared, and the predictive value of these indicators for TAO patients at different activity stages was investigated.

RESULTS:A total of 77 TAO patients were enrolled in this study, including 38 males and 39 females, with ages ranging from 28 to 70 y(mean age 49.5±6.9 y). There were 77 cases in the control group, including 40 males and 37 females, with ages ranging from 26 to 70 y(mean age 49.0±7.3 y). There was no significant difference in gender and age between the two groups(both P>0.05). The quantitative analysis of left eye, right eye, and binocular CT orbital structure in TAO group patients showed significantly higher indicators than the control group(all P<0.001), and the CAS score of TAO group was 3.94±1.51 points. The CAS score was positively correlated with various indicators of CT orbital structure quantitative analysis(all P<0.05). According to the CAS score results, 14 cases(28 eyes)of TAO patients with a CAS score<3 were classified as inactive phase, including 8 males and 6 females, with an average age of 43.79±9.58 y. A total of 63 cases(126 eyes)with a CAS score of ≥3 was classified as active phase, including 30 males and 33 females, with an average age of 50.78±5.47 y. There was no significant difference in gender among TAO patients with different active phases(P=0.519), but there was a significant difference in age(P<0.001). The quantitative indicators of CT orbital structure in inactive patients were significantly lower than those in active patients(P<0.05). Finally, the superior rectus muscle, age, and degree of protrusion were selected to be included in the Logistic regression model. The analysis results showed that there was a correlation between the superior rectus muscle index, degree of protrusion and TAO activity phase(P<0.05), while age, and TAO activity phase showed no significant correlation(P>0.05). The ROC curve analysis results showed that the area under the curve(AUC)was 0.863, the standard error was 0.063, P<0.001, and the 95% confidence interval(95% CI)of AUC was 0.740-0.985. The sensitivity of the model prediction was 73.0%, the specificity was 92.9%, and the Youden index was 0.659. The prediction accuracy was 97.9%, the recall rate was 73.0%, and the F1 value was 0.836. The predicted optimal critical value was 0.857. The predicted probability was 0.74.

CONCLUSION:Quantitative CT analysis of orbital structures can be used to assess disease severity in TAO patients.