Diagnostic Efficacy of Quantitative Computed Tomography in CTD-ILA/ILD
-
摘要:
目的:定量CT在结缔组织病相关间质性肺异常(CTD-ILA)和间质性肺疾病(CTD-ILD)中的诊断效能,建立基于定量CT的CTD患者筛查方法。方法:纳入CTD-ILD患者140例、CTD-ILA患者33例及对照组109例,使用3D-Slicer获得定量指标。结果:各组间定量CT指标均存在差异;ROC分析显示,F%、GGO%、SD及Kurtosis是鉴别对照组与CTD-ILA/ILD的敏感指标,其中SD在早期诊断CTD-ILA(AUC=0.862)、CTD-ILD(AUC=0.923)时表现最佳,进一步区分CTD-ILA与CTD-ILD时,SD(AUC=0.649)和F%(AUC=0.617)展现出较强区分能力;多元逐步logistic回归分析显示,F%、GGO%、SD和Kurtosis在区分对照组与CTD-ILA/ILD时具有统计学意义。结论:定量CT对于CTD-ILA/ILD早期诊断具有重要意义,基于定量CT构建CTD筛查流程有助实现患者精准管理。
Abstract:Objective: The aim of this study is to evaluate the diagnostic efficacy of quantitative computed tomography (CT) in differentiating between connective tissue disease-associated interstitial lung abnormalities (CTD-ILA) and connective tissue disease-associated interstitial lung disease (CTD-ILD), as well as to establish a screening protocol for connective tissue disease (CTD) patients based on quantitative CT. Methods: A total of 140 patients with CTD-ILD, 33 patients with CTD-ILA, and 109 healthy controls were enrolled. Quantitative indices were obtained using the 3D-Slicer software. Results: Significant differences in quantitative CT indices are observed among the groups. ROC analysis shows that the F%, GGO%, SD, and kurtosis are sensitive indicators for differentiating the control group from those with CTD-ILA/ILD. Notably, the best SD is demonstrated in the early diagnosis of both the CTD-ILA (AUC=0.862) and CTD-ILD (AUC=0.923) groups. Further distinguishing between CTD-ILA and CTD-ILD shows the strong discriminatory ability of the SD (AUC=0.649) and F% (AUC=0.617). Multivariable stepwise logistic regression analysis shows that F%, GGO%, SD, and kurtosis are statistically significant in differentiating the control group from the CTD-ILA/ILD groups. Conclusion: Quantitative CT is promising for the early diagnosis of CTD-ILA/ILD. Establishing a CTD screening protocol based on quantitative CT can facilitate precise patient management.
-
间质性肺异常(interstitial lung anomaly,ILA)是在轻度症状或轻微肺功能损害患者中偶然发现的,提示早期间质性肺疾病(interstitial lung disease,ILD)[1]。Fleischner协会提出,ILA是在无疑似ILD患者中发现与ILD相符的特异性CT表现,并建议对其进行评估[2]。研究表明,ILA很常见[3],且与死亡率增加、肺癌风险和癌症治疗相关并发症显著相关[4-5]。Cedars-Sinai等[6]通过对类风湿关节炎(rheumatoid arthritis,RA)患者进行纵向随访,发现基线时存在ILA的患者1年后出现影像学进展。因此,有必要对ILA进行早期诊断和适当管理。
ILD是结缔组织病(connective tissue disease,CTD)最常见的肺部并发症[7],通常与不良结局和早期死亡率有关[8]。由于CTD-ILD患者可能表现为无或咳嗽等非特异性症状[9],通常在病程晚期才被诊断[10],且与其他疾病重叠,导致ILD的诊断经常被延迟和低估[11],使患者预后恶化[12]。然而,抗纤维化治疗已被证明可降低ILD患者的死亡率和急性加重的风险[13],延缓肺纤维化进展[14]。因此,保持对ILD的高度怀疑,并尽早识别和治疗有助于改善患者的临床结局[15-16]。
目前,高分辨率计算机断层扫描(high-resolution CT,HRCT)是诊断ILD的金标准[17],也是筛查ILA的重要工具[2]。然而,肉眼识别每个肺区5%的肺异常具有挑战性[18],且早期ILA患者与其他肺部病变(如重力依赖型肺不张、局灶胸椎旁纤维化等[19])相似。而定量CT可克服这些局限性,在ILA评估中发挥着重要作用[20]。该技术不仅能量化不同影像表现(如磨玻璃影、纤维化等)的占比,还能评估肺部整体损伤(如平均肺密度等)情况,并且能够早期发现病变,为疾病评估和早期诊断提供重要依据。
然而,针对CTD患者的ILA的研究仍较少。本研究纳入CTD-ILA及CTD-ILD患者,通过定量CT评估患者肺损伤情况,探讨定量CT指标对于CTD-ILA、CTD-ILD的预测作用及诊断效能,建立基于定量CT的CTD患者筛查方法,为早期诊断与个性化管理提供参考依据。
1. 资料与方法
1.1 一般资料
选取延安大学附属医院2018年1月至2024年11月CTD-ILA患者33例、CTD-ILD患者140例,对照组109例。
CTD-ILA/ILD纳入标准:类风湿关节炎(rheumatoid arthritis,RA)、系统性硬化症(systemic sclerosis,SSc)、皮肌炎(dermatomyositis,DM)、系统性红斑狼疮(systemic lupus erythematosus,SLE)、抗中性粒细胞胞浆抗体(anti-neutrophil cytoplasmic antibody,ANCA)相关血管炎的诊断分别符合2010年[21]、2013年[22]、2017年[23]、2019年[24]、2022年[25]美国风湿病学会指南,干燥综合征(sjögren’s syndrome,SS)、弥漫性结缔组织病(mixed connective tissue disease,MCTD)诊断符合2016年[26]、2019年[27]美欧共识小组指南,ILA的定义符合2020年Fleischner协会指南[2],ILD的定义符合2013年英国胸科协会指南[28]。ILA和ILD示例图如图1所示。
CTD-ILA/ILD排除标准:无薄层高分辨率CT;胸部CT伪影大;3D slicer无法进行分割;存在胸部肿瘤、感染[29-30]、肺气肿、胸腔积液、胸部手术史。对照组的纳入标准是符合CTD诊断但并未患有ILD的人群,其排除标准与CTD-ILD/ILD患者的排除标准一致。
1.2 HRCT检查及定量分析
使用128层螺旋CT扫描仪(上海联影UCT-760)进行检查,受试者取仰卧位,于深吸气末扫描,扫描范围为肺尖至肺底膈面。扫描参数:管电压120 kV、管电流自动调节,扫描层厚5 mm,使用1 mm标准算法重建。
由两名经验丰富的放射科主任医生判断患者胸部CT质量并对其进行影像分型。定量CT指标使用3D-Slicer(5.6.2版本,http://www.slicer.org)获得。使用密度直方图和密度阈值法得到定量CT指标。具体步骤如图2所示。密度阈值法的阈值设置:全肺像素阈值设定为 −
1024 HU至 −200 HU,并进一步细分为正常肺组织(−950 HU至 −700 HU)、磨玻璃密度(−700 HU至 −500 HU)和纤维化区域(−500 HU至 −200 HU)。通过计算这些区域像素与全肺像素的比例,得到正常肺区域(NL%)、磨玻璃密度区域(GGO%)和纤维化区域(F%)的百分比,以及异常病变区域(AA%)的百分比(AA%为GGO%与F%之和),分割结果如图3所示。
1.3 统计学方法
采用Shapiro-Wilk W进行正态检验。正态资料以(均数±标准差)表示,非正态资料以中位数(四分位数间距)表示,分类变量以频率(百分比(%))表示。组间差异用单因素方差分析或Kruskal-Wallis H检验分析,采用Bonferroni法进行事后比较。
应用受试者工作特征曲线(receiver operating characteristic,ROC)评估肺定量CT指标在区分对照组与CTD-ILA、对照组与CTD-ILD、CTD-ILA与CTD-ILD的诊断效能,使用多因素多元逐步logistic回归筛选对预测CTD-ILA及CTD-ILD有意义指标。
使用SPSS 25.0进行统计分析,以P<0.05为差异具有统计学意义。
2. 结果
2.1 患者基线特征
本研究共纳入282名患者,包括对照组109例,CTD-ILD组140例,CTD-ILA组33例。在性别、年龄和CTD类型方面,各组间存在显著差异。而病程、BMI以及CT分型方面无显著差异。
表 1 患者临床资料表Table 1. Demographic and clinical characteristics of study participants项目 组别 统计检验 对照组(n=109) CTD-ILA(n=33) CTD-ILD(n=140) F/H p 性别(%) 14.52 0.001 男 19(17.3) 14(42.4) 53(37.9) 女 90(81.8) 19(57.6) 87(62.1) 年龄 40(19) 64(11) 62.5(14) 104.52 0.000 BMI 22.54(2.81) 21.89(3.08) 22.34(4.39) 0.29 0.864 病程 3(7) 5(11.25) 4(10.75) 3.25 0.187 CTD类型(%) 9.66 0.008 类风湿关节炎 63(57.3) 22(66.7) 67(47.9) 系统性红斑狼疮 24(21.8) 5(15.2) 10(7.1) 系统性硬化症 2(1.8) 2(6.1) 15(10.7) 干燥综合症 9(8.2) 2(6.1) 16(11.4) 皮肌炎 − 2(6.1) 8(5.7) 弥漫性结缔组织病 7(6.4) − 15(10.7) ANCA相关血管炎 4(3.6) 2(6.1) 9(6.4) CT分型(%) 1.12 0.291 普通型间质性肺炎 − 7(21.2) 41(29.3) 非特异性间质性肺炎 − 18(54.5) 71(50.7) 淋巴细胞性间质性肺炎 − 4(12.1) 22(15.7) 机化性肺炎 − 4(12.1) 6(4.3) 2.2 各组间定量指标差异
定量CT指标分析显示,各组间定量CT指标均存在统计学差异。在密度阈值法中,从无ILD、ILA到ILD,NL%逐渐下降,而GGO%、F%、AA%则逐渐上升。
在直方图分析法中,平均肺密度(mean lung density,MLD)、高衰减区(high attenuation area,HAA)以及标准差(standard deviation,SD)均呈现上升趋势,而峰度(Kurtosis)和偏度(Skewness)则逐渐减小。
表 2 各组间定量CT指标差异Table 2. Intergroup differences in quantitative CT metrics among controls, CTD-ILA, and CTD-ILD cohorts项目 组别 统计检验 对照组(n=109) CTD-ILA(n=22) CTD-ILD(n=140) F/H p NL% 74(7) 68.0(7.5)* 65.5(9)* 89.82 0.000 GGO% 5.7(3.7) 11.3(8.75)* 12.65(9.1)* 76.00 0.000 F% 2.9(1.25) 5.2(3.4)* 6.6(4.98)* 113.72 0.000 AA% 8.8(4.9) 16.40(11.80)* 19.35(12.67)* 89.51 0.000 HAA 4.04(2.18) 9.25±5.26* 9.44(6.92)* 114.05 0.000 MLD −830.16(49.19) −777.08(59.02)* −768.19(93.69)* 70.18 0.000 SD 179.94(16.38) 205.52(25.83)* 220.19(41.43)* 139.60 0.000 Kurtosis 13.62±4.20 6.77(4.91)* 5.28(5.64)* 122.59 0.000 Skewness 3.27(0.71) 2.34(0.74)* 2.19±0.60* 115.41 0.000 注:*表示与对照组相比P<0.05。NL%为正常肺区域的百分比;GGO%为磨玻璃密度区域的百分比;F%为纤维化区域的百分比;AA%为异常病变区域的百分比;HAA为高衰减区;MLA为平均肺衰减;SD为标准差;Kurtosis为峰值;Skewness为偏度。 2.3 定量CT指标区分ILA与ILD ROC曲线
ROC曲线分析结果显示,SD、HAA、Kurtosis和F%在区分对照组与CTD-ILA(表3和图4)、对照组与CTD-ILD(表4和图5)方面表现优异。其中,SD的曲线下面积(AUC)最大,在区分对照组与CTD-ILA(AUC=0.862)和对照组与CTD-ILD(AUC=0.923)时均表现出较高的诊断效能。在密度阈值法中,F%区分两者AUC值最大。
表 3 定量CT指标区分对照组与CTD-ILA的ROC曲线分析结果Table 3. ROC curve analysis of quantitative CT metrics for discriminating control groups from CTD-ILA patients项目 AUC1 P 最佳截断值 灵敏度 特异度 约登指数 NL% 0.770 0.001 < 71.50 0.758 0.743 0.501 GGO% 0.774 0.001 > 8.000 0.788 0.734 0.522 F% 0.814 0.001 > 4.050 0.758 0.807 0.565 AA% 0.785 0.001 > 12.20 0.788 0.762 0.549 HAA 0.830 0.001 > 6.121 0.788 0.835 0.623 MLD 0.763 0.001 > −800.9 0.697 0.798 0.495 SD 0.862 0.001 > 191.2 0.849 0.817 0.665 Kurtosis 0.821 0.001 < 9.615 0.727 0.844 0.571 Skewness 0.816 0.001 < 2.795 0.788 0.817 0.604 注:NL% 为正常肺区域的百分比;GGO%为磨玻璃密度区域的百分比;F%为纤维化区域的百分比;AA%为异常病变区域的百分比;HAA为高衰减区;MLA为平均肺衰减;SD为标准差; Kurtosis为峰值; Skewness为偏度。 表 4 定量CT指标区分对照组与CTD-ILD的ROC曲线分析结果Table 4. ROC curve analysis of quantitative CT metrics for differentiating control groups from CTD-ILD patients项目 AUC2 P 最佳截断值 灵敏度 特异度 约登指数 NL% 0.844 0.000 < 71.50 0.829 0.743 0.572 GGO% 0.815 0.000 > 8.350 0.771 0.752 0.524 F% 0.886 0.000 > 4.050 0.727 0.807 0.535 AA% 0.843 0.000 > 11.30 0.857 0.716 0.573 HAA 0.885 0.000 > 6.205 0.727 0.844 0.571 MLD 0.801 0.000 > −798.2 0.636 0.826 0.462 SD 0.923 0.000 > 191.2 0.864 0.817 0.680 Kurtosis 0.896 0.000 < 11.74 0.818 0.706 0.525 Skewness 0.889 0.000 < 2.795 0.727 0.817 0.544 注:NL%为正常肺区域的百分比;GGO%为磨玻璃密度区域的百分比;F%为纤维化区域的百分比;AA%为异常病变区域的百分比;HAA为高衰减区;MLA为平均肺衰减;SD为标准差; Kurtosis为峰值; Skewness为偏度。 在区分CTD-ILA与CTD-ILD的ROC曲线分析(表5和图6),SD的AUC最大为(AUC=0.649 ),其次为Kurtosis(AUC=0.638),F%在密度阈值法中AUC最大为0.617。
表 5 定量CT指标区分CTD-ILA与CTD-ILD的ROC曲线分析结果Table 5. ROC curve analysis of quantitative CT metrics for differentiating CTD-ILA from CTD-ILD项目 AUC3 P 最佳截断值 灵敏度 特异度 约登指数 NL% 0.600 0.074 < 63.50 0.379 0.788 0.167 GGO% 0.544 0.435 > 12.10 0.536 0.576 0.112 F% 0.617 0.037 > 5.450 0.650 0.576 0.226 AA% 0.574 0.187 > 16.85 0.600 0.576 0.176 HAA 0.585 0.128 > 10.06 0.486 0.727 0.213 MLD 0.568 0.224 > −758.5 0.429 0.758 0.186 SD 0.649 0.008 > 219.1 0.529 0.758 0.286 Kurtosis 0.638 0.014 < 5.095 0.479 0.849 0.327 Skewness 0.614 0.042 < 1.932 0.364 0.879 0.243 注:NL%为正常肺区域的百分比;GGO%为磨玻璃密度区域的百分比;F%为纤维化区域的百分比;AA%为异常病变区域的百分比;HAA为高衰减区;MLA为平均肺衰减;SD为标准差; Kurtosis为峰值; Skewness为偏度。 2.4 定量CT指标预测ILA与ILD 多元logistic回归分析
将NL%、GGO%、F%、AA%、HAA、MLD、SD、Kurtosis、Skewness作为自变量,分组结果(ILA/ILD)为因变量,进行多因素多元逐步logistic回归,以无ILD的对照组为参考进行对比分析。
结果显示,在区分对照组与ILD及ILA时,F%(β=0.620,P=0.006,OR=1.895;β=0.854,P=0.000,OR=2.349)、GGO%(β=−0.302,P=0.002,OR=0.739;β=−0.454,P=0.000,OR=0.635)、SD(β=0.026,P=0.009,OR=1.016;β=0.016,P=0.000,OR=1.026)、Kurtosis(β=−0.370,P=0.000,OR=0.691;β=−0.439,P=0.000,OR=0.645)具有统计学差异。
表 6 定量CT指标预测ILA与ILD多因素多元logistic回归结果Table 6. Multivariate logistic regression analysis of quantitative CT metrics in predicting ILA vs. ILD项目 β BE wald P OR 95% CI 下限 上限 ILA F% 0.620 0.225 7.629 0.006 1.859 1.197 2.887 GGO% −0.302 0.098 9.546 0.002 0.739 0.610 0.895 SD 0.015 0.006 6.784 0.009 1.016 1.004 1.027 Kurtosis −0.370 0.069 28.727 0.000 0.691 0.603 0.791 ILD F% 0.854 0.209 16.666 0.000 2.349 1.559 3.540 GGO% −0.454 0.092 24.385 0.000 0.635 0.530 0.760 SD 0.026 0.005 23.855 0.000 1.026 1.016 1.037 Kurtosis −0.439 0.062 50.962 0.000 0.645 0.571 0.727 注:NL%为正常肺区域的百分比;GGO%为磨玻璃密度区域的百分比;F%为纤维化区域的百分比;AA%为异常病变区域的百分比;HAA为高衰减区;MLA为平均肺衰减;SD为标准差; Kurtosis为峰值; Skewness为偏度。 3. 讨论
ILA及ILD均与CTD患者不良预后和早期死亡率密切相关,早期识别和管理对改善患者结局至关重要。但目前临床筛查仍依赖视觉评估,易导致早期病变漏诊。与既往研究相比,本研究系统量化了从正常肺组织、ILA到ILD的连续影像演变规律,为疾病的发展提供了客观证据;其次,构建基于定量CT的CTD患者筛查方法,为建立风险分层提供依据;最后,使用开源分析软件,免费、省时,有利于技术推广与普及。
从无ILD的CTD对照组、CTD-ILA到CTD-ILD,NL%、峰度(Kurtosis)和偏度(Skewness)逐渐下降,而GGO%、F%、AA%、MLD、HAA以及SD则逐渐上升,与Ahn等[31]及杨凯等[32]的研究一致,表明ILA是ILD的早期阶段。
定量CT指标的ROC曲线和Logistic回归分析表明,F%、GGO%、SD和Kurtosis是区分对照组与CTD-ILA/ILD的敏感指标。在密度阈值法,F%在区分对照组与CTD-ILA(AUC=0.814)和CTD-ILD(AUC=0.886)时表现最佳,提示其在早期诊断ILA和ILD方面具有重要意义。与Zhang等[33]研究结果一致,F%同样被证实为识别健康对照与特发性肺间质纤维化(IPF)的最佳指标(AUC=0.962)。但本研究F%的AUC值低于其在IPF中的表现,这提示炎症主导的CTD-ILD与纤维化驱动的IPF存在本质差异[34],这说明定量指标筛选应结合患者病因。
此外,磨玻璃密度区域百分比(GGO%)在诊断ILA及ILD方面同样具有重要价值(β1=−0.302,P=0.002,OR=0.739;β2=−0.454,P=0.000,OR=0.635)。这与马震忠等[35]发现皮肌炎/多发性肌炎相关ILD(DM/PM-ILD)患者HRCT表现以磨玻璃影为主(87.2%,123/141)一致,徐光兴等[36]证实GGO%在鉴别DM/PM-ILD与健康对照组时具有较高的诊断效能(AUC=0.91),但DM/PM-ILD患者的影像表现以非特异性间质性肺炎(NSIP)为主[37],而本研究还纳入了30%左右以普通型间质性肺炎(UIP)表现为主的患者,这进一步证明GGO%在鉴别不同类型ILD时的广泛适用性。
在直方图分析法中,标准差(SD)和峰度(Kurtosis)是区分正常组和CTD-ILA/ILD有意义的指标。SD的曲线下面积最大(AUC1=0.862,AUC2=0.923),Kurtosis区分ILD(AUC2=0.896)的AUC值仅此于SD(AUC2=0.923),与Guisado-Vasco等[38]研究一致。研究表明SD还是区分局限性和弥漫性SSc-ILD和SS-ILD的最佳鉴别参数[39-40],进一步证明SD对于判断疾病严重程度也具有重要价值。
本研究还建立区分ILA和ILD的ROC曲线,结果显示SD的AUC最大(AUC=0.649),SD>219.1为区分ILA和ILD的最佳截断值。在密度阈值法,F%的AUC最大(AUC=0.617),F%>5.450为区分ILA和ILD的最佳截断值。然而,定量CT结果仅供参考,需结合临床症状、肺功能检查及随访进行综合评估。
基于当前证据,我们提出CTD肺部评估的大致路径:初筛阶段采用SD、F%、GGO%、和Kurtosis排除高危人群;对于异常患者采用SD和F%判断患者是患有ILA还是ILD。对于ILA患者,应定期随访监测病情变化;而ILD患者,则评估病情程度,尽早治疗改善预后。
尽管高衰减区(high attenuation area,HAA)未被纳入回归模型,但其在区分对照组与ILA患者时表现较好(AUC=0.830)。既往研究也表明,HAA在无症状类风湿关节炎(RA)识别ILD风险的能力最强[41],不同肺区域HAA与肺重塑的生物标志物、ILA的风险和全因死亡率相关[42],呼气HAA被认为是所有直方图分析指标中ILD严重程度的最佳预测因子[43],均提示HAA对于早期诊断的价值。此外,Shiraishi等[44]纵向评估COPD患者ILA,发现ILDvol%(磨玻璃影、网状影和蜂窝影的体积之和)与ILA 的发生相关,是识别和监测COPD患者ILA的可重复方法。这与我们的AA%指标类似,但AA%区分对照组与CTD-ILA/ILD AUC值较小,可能是3D slicer无法区分蜂窝影与肺气肿的局限性,未来需通过深度学习算法优化分割精度。
本研究的局限性。①样本量较小,尤其是CTD-ILA组。由于前期研究重点是CTD-ILD,对ILA人群关注较少,但本研究仍为该领域提供了有价值的依据,未来进一步扩大样本量。②本研究在纳入年龄和性别因素时,发现模型参数并未显著优化,因此未将其纳入最终分析。3D slicer目前无法区分肺气肿与蜂窝结构,因此纳入人群时已排除有肺气肿的患者。③本研究主要聚焦于定量CT的诊断价值,而忽略了血清学指标、肺功能等的诊断意义,后续研究将纳入临床指标。
综上所述,本研究不仅验证了定量CT在CTD-ILD/ILA早期诊断中的价值,还提出基于定量CT的CTD患者的筛查方式,为患者精准管理提供有力支持。未来需在大样本中验证其适用性,并探索其与临床指标的协同作用。
-
表 1 患者临床资料表
Table 1 Demographic and clinical characteristics of study participants
项目 组别 统计检验 对照组(n=109) CTD-ILA(n=33) CTD-ILD(n=140) F/H p 性别(%) 14.52 0.001 男 19(17.3) 14(42.4) 53(37.9) 女 90(81.8) 19(57.6) 87(62.1) 年龄 40(19) 64(11) 62.5(14) 104.52 0.000 BMI 22.54(2.81) 21.89(3.08) 22.34(4.39) 0.29 0.864 病程 3(7) 5(11.25) 4(10.75) 3.25 0.187 CTD类型(%) 9.66 0.008 类风湿关节炎 63(57.3) 22(66.7) 67(47.9) 系统性红斑狼疮 24(21.8) 5(15.2) 10(7.1) 系统性硬化症 2(1.8) 2(6.1) 15(10.7) 干燥综合症 9(8.2) 2(6.1) 16(11.4) 皮肌炎 − 2(6.1) 8(5.7) 弥漫性结缔组织病 7(6.4) − 15(10.7) ANCA相关血管炎 4(3.6) 2(6.1) 9(6.4) CT分型(%) 1.12 0.291 普通型间质性肺炎 − 7(21.2) 41(29.3) 非特异性间质性肺炎 − 18(54.5) 71(50.7) 淋巴细胞性间质性肺炎 − 4(12.1) 22(15.7) 机化性肺炎 − 4(12.1) 6(4.3) 表 2 各组间定量CT指标差异
Table 2 Intergroup differences in quantitative CT metrics among controls, CTD-ILA, and CTD-ILD cohorts
项目 组别 统计检验 对照组(n=109) CTD-ILA(n=22) CTD-ILD(n=140) F/H p NL% 74(7) 68.0(7.5)* 65.5(9)* 89.82 0.000 GGO% 5.7(3.7) 11.3(8.75)* 12.65(9.1)* 76.00 0.000 F% 2.9(1.25) 5.2(3.4)* 6.6(4.98)* 113.72 0.000 AA% 8.8(4.9) 16.40(11.80)* 19.35(12.67)* 89.51 0.000 HAA 4.04(2.18) 9.25±5.26* 9.44(6.92)* 114.05 0.000 MLD −830.16(49.19) −777.08(59.02)* −768.19(93.69)* 70.18 0.000 SD 179.94(16.38) 205.52(25.83)* 220.19(41.43)* 139.60 0.000 Kurtosis 13.62±4.20 6.77(4.91)* 5.28(5.64)* 122.59 0.000 Skewness 3.27(0.71) 2.34(0.74)* 2.19±0.60* 115.41 0.000 注:*表示与对照组相比P<0.05。NL%为正常肺区域的百分比;GGO%为磨玻璃密度区域的百分比;F%为纤维化区域的百分比;AA%为异常病变区域的百分比;HAA为高衰减区;MLA为平均肺衰减;SD为标准差;Kurtosis为峰值;Skewness为偏度。 表 3 定量CT指标区分对照组与CTD-ILA的ROC曲线分析结果
Table 3 ROC curve analysis of quantitative CT metrics for discriminating control groups from CTD-ILA patients
项目 AUC1 P 最佳截断值 灵敏度 特异度 约登指数 NL% 0.770 0.001 < 71.50 0.758 0.743 0.501 GGO% 0.774 0.001 > 8.000 0.788 0.734 0.522 F% 0.814 0.001 > 4.050 0.758 0.807 0.565 AA% 0.785 0.001 > 12.20 0.788 0.762 0.549 HAA 0.830 0.001 > 6.121 0.788 0.835 0.623 MLD 0.763 0.001 > −800.9 0.697 0.798 0.495 SD 0.862 0.001 > 191.2 0.849 0.817 0.665 Kurtosis 0.821 0.001 < 9.615 0.727 0.844 0.571 Skewness 0.816 0.001 < 2.795 0.788 0.817 0.604 注:NL% 为正常肺区域的百分比;GGO%为磨玻璃密度区域的百分比;F%为纤维化区域的百分比;AA%为异常病变区域的百分比;HAA为高衰减区;MLA为平均肺衰减;SD为标准差; Kurtosis为峰值; Skewness为偏度。 表 4 定量CT指标区分对照组与CTD-ILD的ROC曲线分析结果
Table 4 ROC curve analysis of quantitative CT metrics for differentiating control groups from CTD-ILD patients
项目 AUC2 P 最佳截断值 灵敏度 特异度 约登指数 NL% 0.844 0.000 < 71.50 0.829 0.743 0.572 GGO% 0.815 0.000 > 8.350 0.771 0.752 0.524 F% 0.886 0.000 > 4.050 0.727 0.807 0.535 AA% 0.843 0.000 > 11.30 0.857 0.716 0.573 HAA 0.885 0.000 > 6.205 0.727 0.844 0.571 MLD 0.801 0.000 > −798.2 0.636 0.826 0.462 SD 0.923 0.000 > 191.2 0.864 0.817 0.680 Kurtosis 0.896 0.000 < 11.74 0.818 0.706 0.525 Skewness 0.889 0.000 < 2.795 0.727 0.817 0.544 注:NL%为正常肺区域的百分比;GGO%为磨玻璃密度区域的百分比;F%为纤维化区域的百分比;AA%为异常病变区域的百分比;HAA为高衰减区;MLA为平均肺衰减;SD为标准差; Kurtosis为峰值; Skewness为偏度。 表 5 定量CT指标区分CTD-ILA与CTD-ILD的ROC曲线分析结果
Table 5 ROC curve analysis of quantitative CT metrics for differentiating CTD-ILA from CTD-ILD
项目 AUC3 P 最佳截断值 灵敏度 特异度 约登指数 NL% 0.600 0.074 < 63.50 0.379 0.788 0.167 GGO% 0.544 0.435 > 12.10 0.536 0.576 0.112 F% 0.617 0.037 > 5.450 0.650 0.576 0.226 AA% 0.574 0.187 > 16.85 0.600 0.576 0.176 HAA 0.585 0.128 > 10.06 0.486 0.727 0.213 MLD 0.568 0.224 > −758.5 0.429 0.758 0.186 SD 0.649 0.008 > 219.1 0.529 0.758 0.286 Kurtosis 0.638 0.014 < 5.095 0.479 0.849 0.327 Skewness 0.614 0.042 < 1.932 0.364 0.879 0.243 注:NL%为正常肺区域的百分比;GGO%为磨玻璃密度区域的百分比;F%为纤维化区域的百分比;AA%为异常病变区域的百分比;HAA为高衰减区;MLA为平均肺衰减;SD为标准差; Kurtosis为峰值; Skewness为偏度。 表 6 定量CT指标预测ILA与ILD多因素多元logistic回归结果
Table 6 Multivariate logistic regression analysis of quantitative CT metrics in predicting ILA vs. ILD
项目 β BE wald P OR 95% CI 下限 上限 ILA F% 0.620 0.225 7.629 0.006 1.859 1.197 2.887 GGO% −0.302 0.098 9.546 0.002 0.739 0.610 0.895 SD 0.015 0.006 6.784 0.009 1.016 1.004 1.027 Kurtosis −0.370 0.069 28.727 0.000 0.691 0.603 0.791 ILD F% 0.854 0.209 16.666 0.000 2.349 1.559 3.540 GGO% −0.454 0.092 24.385 0.000 0.635 0.530 0.760 SD 0.026 0.005 23.855 0.000 1.026 1.016 1.037 Kurtosis −0.439 0.062 50.962 0.000 0.645 0.571 0.727 注:NL%为正常肺区域的百分比;GGO%为磨玻璃密度区域的百分比;F%为纤维化区域的百分比;AA%为异常病变区域的百分比;HAA为高衰减区;MLA为平均肺衰减;SD为标准差; Kurtosis为峰值; Skewness为偏度。 -
[1] TOMASSETTI S, POLETTI V, RAVAGLIA C, et al. Incidental discovery of interstitial lung disease: Diagnostic approach, surveillance and perspectives[J]. European Respiratory Review, 2022, 31(164): 210206. DOI: 10.1183/16000617.0206-2021.
[2] HATABU H, HUNNINGHAKE G M, RICHELDI L, et al. Interstitial lung abnormalities detected incidentally on CT: Position Paper from the Fleischner Society[J]. The Lancet Respiratory Medicine, 2020, 8(7): 726-737. DOI: 10.1016/S2213-2600(20)30168-5.
[3] GARCíA MULLOR M M, ARENAS-JIMéNEZ J J, UREñA VACAS A, et al. Prevalence and prognostic meaning of interstitial lung abnormalities in remote CT scans of patients with interstitial lung disease treated with antifibrotic therapy[J]. Radiología (English Edition), 2024, 66: S10-S23. DOI: 10.1016/j.rxeng.2023.03.006.
[4] SEOK J, PARK S, YOON E C, et al. Clinical outcomes of interstitial lung abnormalities: A systematic review and meta-analysis[J/OL]. Scientific Reports, 2024, 14(1): 7330. DOI: 10.1038/s41598-024-57831-3.
[5] AXELSSON G T, PUTMAN R K, ASPELUND T, et al. The associations of interstitial lung abnormalities with cancer diagnoses and mortality[J]. European Respiratory Journal, 2020, 56(6): 1902154. DOI: 10.1183/13993003.02154-2019.
[6] DONG H, JULIEN P J, DEMORUELLE M K, et al. Interstitial lung abnormalities in patients with early rheumatoid arthritis: A pilot study evaluating prevalence and progression[J]. European Journal of Rheumatology, 2018, 6(4): 193-198. DOI: 10.5152/eurjrheum.2019.19044.
[7] JEE A S, SHEEHY R, HOPKINS P, et al. Diagnosis and management of connective tissue disease‐associated interstitial lung disease in Australia and New Zealand: A position statement from the Thoracic Society of Australia and New Zealand[J]. Respirology, 2021, 26(1): 23-51. DOI: 10.1111/resp.13977.
[8] HU Z, WANG H, HUANG J, et al. Cardiovascular disease in connective tissue disease-associated interstitial lung disease: A systematic review and meta-analysis of observational studies[J]. Autoimmunity Reviews, 2024, 23(10): 103614. DOI: 10.1016/j.autrev.2024.103614.
[9] JEGANATHAN N, SATHANANTHAN M. Connective tissue disease-related interstitial lung disease: Prevalence, patterns, predictors, prognosis, and treatment[J]. Lung, 2020, 198(5): 735-759. DOI: 10.1007/s00408-020-00383-w.
[10] SPAGNOLO P, RYERSON C J, PUTMAN R, et al. Early diagnosis of fibrotic interstitial lung disease: Challenges and opportunities[J]. The Lancet Respiratory Medicine, 2021, 9(9): 1065-1076. DOI: 10.1016/S2213-2600(21)00017-5.
[11] PRITCHARD D, ADEGUNSOYE A, LAFOND E, et al. Diagnostic test interpretation and referral delay in patients with interstitial lung disease[J]. Respiratory Research, 2019, 20(1): 253. DOI: 10.1186/s12931-019-1228-2.
[12] CANO-JIMéNEZ E, VáZQUEZ RODRíGUEZ T, MARTíN-ROBLES I, et al. Diagnostic delay of associated interstitial lung disease increases mortality in rheumatoid arthritis[J]. Scientific Reports, 2021, 11(1): 9184. DOI: 10.1038/s41598-021-88734-2.
[13] PETNAK T, LERTJITBANJONG P, THONGPRAYOON C, et al. Impact of antifibrotic therapy on mortality and acute exacerbation in idiopathic pulmonary fibrosis[J]. Chest, 2021, 160(5): 1751-1763. DOI: 10.1016/j.chest.2021.06.049.
[14] GHAZIPURA M, MAMMEN M J, HERMAN D D, et al. Nintedanib in progressive pulmonary fibrosis: A systematic review and meta-analysis[J]. Annals of the American Thoracic Society, 2022, 19(6): 1040-1049. DOI: 10.1513/AnnalsATS.202103-343OC.
[15] DUBEY S, WOODHEAD F. Survival differences in rheumatoid arthritis interstitial lung disease and idiopathic pulmonary fibrosis may be explained by delays in presentation: Results from multivariate analysis in a monocentric UK study[J]. Rheumatology International, 2023, 44(1): 99-105. DOI: 10.1007/s00296-023-05505-0.
[16] HEWITT R J, BARTLETT E C, GANATRA R, et al. Lung cancer screening provides an opportunity for early diagnosis and treatment of interstitial lung disease[J]. Thorax, 2022, 77(11): 1149-1151. DOI: 10.1136/thorax-2022-219068.
[17] GUIOT J, MIEDEMA J, CORDEIRO A, et al. Practical guidance for the early recognition and follow-up of patients with connective tissue disease-related interstitial lung disease[J]. Autoimmunity Reviews, 2024, 23(6): 103582. DOI: 10.1016/j.autrev.2024.103582.
[18] KIM M S, CHOE J, HWANG H J, et al. Interstitial lung abnormalities (ILA) on routine chest CT: Comparison of radiologists’ visual evaluation and automated quantification[J]. European Journal of Radiology, 2022, 157: 110564. DOI: 10.1016/j.ejrad.2022.110564.
[19] CHAE K J, JIN G Y, GOO J M, et al. Interstitial lung abnormalities: What radiologists should know[J]. Korean Journal of Radiology, 2021, 22(3): 454. DOI: 10.3348/kjr.2020.0191.
[20] 杜雯娟, 赵祥博, 赵海峰, 等. 肺间质异常CT研究进展[J]. 生物医学工程与临床, 2025, 29(1): 129-133. DOI: 10.13339/j.cnki.sglc.20241220.009. [21] ALETAHA D, NEOGI T, SILMAN A J, et al. 2010 Rheumatoid arthritis classification criteria: an American College of Rheumatology/European League Against Rheumatism collaborative initiative[J]. Annals of the Rheumatic Diseases, 2010, 69(9): 1580-1588. DOI: 10.1136/ard.2010.138461.
[22] Van Den HOOGEN F, KHANNA D, FRANSEN J, et al. 2013 classification criteria for systemic sclerosis: An American college of rheumatology/European league against rheumatism collaborative initiative[J]. Annals of the Rheumatic Diseases, 2013, 72(11): 1747-1755. DOI: 10.1136/annrheumdis-2013-204424.
[23] LUNDBERG I E, TJäRNLUND A, BOTTAI M, et al. 2017 European League Against Rheumatism/American College of Rheumatology Classification Criteria for Adult and Juvenile Idiopathic Inflammatory Myopathies and Their Major Subgroups[J]. Arthritis & Rheumatology, 2017, 69(12): 2271-2282. DOI: 10.1002/art.40320.
[24] ARINGER M, COSTENBADER K, DAIKH D, et al. 2019 European League Against Rheumatism/American College of Rheumatology Classification Criteria for Systemic Lupus Erythematosus[J]. Arthritis & Rheumatology, 2019, 71(9): 1400-1412. DOI: 10.1002/art.40930.
[25] SUPPIAH R, ROBSON J C, GRAYSON P C, et al. 2022 American College of Rheumatology/European Alliance of Associations for Rheumatology classification criteria for microscopic polyangiitis[J]. Annals of the Rheumatic Diseases, 2022, 81(3): 321-326. DOI: 10.1136/annrheumdis-2021-221796.
[26] SHIBOSKI C H, SHIBOSKI S C, SEROR R, et al. 2016 American College of Rheumatology/European League Against Rheumatism Classification Criteria for Primary Sjögren’s Syndrome: A Consensus and Data‐Driven Methodology Involving Three International Patient Cohorts[J]. Arthritis & Rheumatology, 2017, 69(1): 35-45. DOI: 10.1002/art.39859.
[27] TANAKA Y, KUWANA M, FUJII T, et al. 2019 Diagnostic criteria for mixed connective tissue disease (MCTD): From the Japan research committee of the ministry of health, labor, and welfare for systemic autoimmune diseases[J]. Modern Rheumatology, 2021, 31(1): 29-33. DOI: 10.1080/14397595.2019.1709944.
[28] TRAVIS W D, COSTABEL U, HANSELL D M, et al. An Official American Thoracic Society/European Respiratory Society Statement: Update of the International Multidisciplinary Classification of the Idiopathic Interstitial Pneumonias[J]. American Journal of Respiratory and Critical Care Medicine, 2013, 188(6): 733-748. DOI: 10.1164/rccm.201308-1483ST.
[29] WIJSENBEEK M, SUZUKI A, MAHER T M. Interstitial lung diseases[J]. The Lancet, 2022, 400(10354): 769-786. DOI: 10.1016/S0140-6736(22)01052-2.
[30] FISCHER A, BOIS R D U. Interstitial lung disease in connective tissue disorders[J]. The Lancet, 2012, 380(9842): 689-698. DOI: 10.1016/S0140-6736(12)61079-4.
[31] AHN Y, LEE S M, CHOI S, et al. Automated CT quantification of interstitial lung abnormality and interstitial lung disease according to the Fleischner Society in patients with resectable lung cancer: Prognostic significance[J]. European Radiology, 2023, 33(11): 8251-8262. DOI: 10.1007/s00330-023-09783-x.
[32] 杨凯, 张静平, 何立宇, 等. 基于定量CT评估多发性肌炎/皮肌炎相关间质性肺病患者肺部改变[J]. 中国临床医学影像杂志, 2024, 35(10): 694-699. YANG K, ZHANG J P, HE L Y, et al. Evaluation of pulmonary changes in patients with polymyositis/dermatomyositis-associated interstitial lung disease based on quantitative CT[J]. Journal of China Clinic Medical Imaging, 2024, 35(10): 694-699. (in Chinese).
[33] ZHANG H, LI X, ZHANG X, et al. Quantitative CT analysis of idiopathic pulmonary fibrosis and correlation with lung function study[J]. BMC Pulmonary Medicine, 2024, 24(1): 437. DOI: 10.1186/s12890-024-03254-9.
[34] JOHANNSON K A, CHAUDHURI N, ADEGUNSOYE A, et al. Treatment of fibrotic interstitial lung disease: Current approaches and future directions[J]. The Lancet, 2021, 398(10309): 1450-1460. DOI: 10.1016/S0140-6736(21)01826-2.
[35] 马震忠, 盛亚丹, 杨凯, 等. 皮肌炎/多发性肌炎相关间质性肺病高分辨率CT特征[J]. CT理论与应用研究(中英文), 2024, 33(4): 497-502. DOI: 10.15953/j.ctta.2023.131. MA Z Z, SHENG Y D, YANG K, et al. HRCT features of dermatomyositis-/polymyositis- associated interstitial lung disease[J]. CT Theory and Applications, 2024, 33(4): 497-502. DOI: 10.15953/j.ctta.2023.131. (in Chinese).
[36] 徐光兴, 俞咏梅 徐亮, 等. 皮肌炎/多发性肌炎并发间质性肺病的CT定量分析与肺功能的相关性研究[J]. 放射学实践, 2023, 38(5): 565-570. XU G X, YU Y M, XU L, et al. Correlation between CT quantitative analysis and pulmonary function of interstitial lung disease in derma-tomyositis/polymyositis[J]. Radiologic Practice, 2023, 38(5): 565-570. (in Chinese).
[37] JEGANATHAN N, SATHANANTHAN M. Connective tissue disease-related interstitial lung disease: Prevalence, patterns, predictors, prognosis, and treatment[J]. Lung, 2020, 198(5): 735-759. DOI: 10.1007/s00408-020-00383-w.
[38] GUISADO-VASCO P, SILVA M, DUARTE-MILLÁN M A, et al. Quantitative assessment of interstitial lung disease in Sjögren’s syndrome[J]. PLoS ONE, 2019, 14(11): e0224772. doi: 10.1371/journal.pone.0224772 GUISADO-VASCO P, SILVA M, DUARTE-MILLÁN M A, et al. Quantitative assessment of interstitial lung disease in Sjögren’s syndrome[J]. PLoS ONE, 2019, 14(11): e0224772. DOI: 10.1371/journal.pone.0224772.
[39] UFUK F, DEMIRCI M, ALTINISIK G. Quantitative computed tomography assessment for systemic sclerosis–related interstitial lung disease: Comparison of different methods[J]. European Radiology, 2020, 30(8): 4369-4380. DOI: 10.1007/s00330-020-06772-2.
[40] UFUK F, DEMIRCI M, ALTINISIK G, et al. Quantitative analysis of Sjogren’s syndrome related interstitial lung disease with different methods[J]. European Journal of Radiology, 2020, 128: 109030. DOI: 10.1016/j.ejrad.2020.109030.
[41] ALEVIZOS M K, DANOFF S K, PAPPAS D A, et al. Assessing predictors of rheumatoid arthritis-associated interstitial lung disease using quantitative lung densitometry[J]. Rheumatology, 2022, 61(7): 2792-2804. DOI: 10.1093/rheumatology/keab828.
[42] CHOI B, KAWUT S M, RAGHU G, et al. Regional distribution of high-attenuation areas on chest computed tomography in the multi-ethnic Study of atherosclerosis[J/OL]. European Respiratory Journal Open Research, 2020, 6(1). [2024-10-16]. https://openres.ersjournals.com/content/6/1/00115-2019. DOI: 10.1183/23120541.00115-2019.
[43] HASAN D, IMAM H, MEGALLY H, et al. The qualitative and quantitative high-resolution computed tomography in the evaluation of interstitial lung diseases[J]. Egyptian Journal of Radiology and Nuclear Medicine, 2020, 51(1): 135. DOI: 10.1186/s43055-020-00254-7.
[44] SHIRAISHI Y, TANABE N, SAKAMOTO R, et al. Longitudinal assessment of interstitial lung abnormalities on CT in patients with COPD using artificial intelligence-based segmentation: A prospective observational study[J]. BMC Pulmonary Medicine, 2024, 24(1): 200. DOI: 10.1186/s12890-024-03002-z.