T细胞线粒体损伤指数在慢性阻塞性肺疾病急性加重风险预测中的价值

doi:10.3969/j.issn.1674-8115.2026.05.009

摘要/Abstract

摘要：

目的·探讨T细胞线粒体损伤指数（mitochondrial damage index，MDI）在慢性阻塞性肺疾病急性加重（acute exacerbation of chronic obstructive pulmonary disease，AECOPD）风险预测中的价值。方法·纳入2023年5月至2024年12月在金华市中心医院呼吸与危重症医学科就诊的慢性阻塞性肺疾病（chronic obstructive pulmonary disease，COPD）患者100例，根据支气管舒张后第一秒用力呼气容积（forced expiratory volume in 1 second，FEV₁）占预计值百分比（FEV₁%pred），将其分为中度COPD组（50%≤FEV₁%pred<80%，n=50）和重度COPD组（30%≤FEV₁%pred<50%，n=50）；另外纳入50例同期健康体检者作为健康对照组。收集所有受试者的基线特征（包括一般临床资料及实验室指标），采用流式细胞术检测3组受试者的外周血T细胞及其亚群（CD3⁺、CD3⁺CD4⁺、CD3⁺CD8⁺）MDI并进行比较分析。对COPD患者随访6个月，以随访期间是否发生急性加重（acute exacerbation，AE）为结局将其分为AE组和非AE组，比较2组患者基线特征的差异。采用LASSO回归对AECOPD的基线预测变量进行筛选，利用单因素及多因素Logistic回归模型建立AECOPD风险的预测模型，并使用ROC曲线、校准曲线和5折交叉验证法验证该模型的性能。结果·COPD患者的T细胞及其亚群MDI异常率均高于健康对照组；其中，重度COPD组患者的CD3⁺CD8⁺ T细胞MDI异常率显著高于中度COPD组和健康对照组（均P<0.05）。随访6个月后，46例COPD患者发生AE。与非AE组相比，AE组患者的FEV₁更低（P<0.001）、COPD评估测试（COPD assessment test，CAT）评分更高（P<0.001）；T细胞及其亚群MDI的各损伤分级分布的组间差异具有统计学意义（均P<0.001），且AE组患者的轻型至重型损伤比例更高。LASSO回归筛选出4个基线预测变量：CD3⁺ T MDI、CD3⁺CD8⁺ T MDI、FEV₁和CAT评分。基于该4个变量构建的AECOPD风险预测模型的AUC为0.848（95%CI 0.772~0.923，P<0.001）、灵敏度为70.0%、特异度为81.0%，校准曲线显示模型拟合良好（P=0.136），5折交叉验证中训练集和验证集准确度分别为0.80±0.03和0.82±0.06。结论·基于CD3⁺ T MDI、CD3⁺CD8⁺ T MDI、FEV₁和CAT评分构建的AECOPD风险预测模型具有较高的预测价值，可为临床早期干预提供参考。

关键词: 慢性阻塞性肺疾病, 急性加重, T细胞及其亚群, 线粒体损伤指数, 预测模型

Abstract:

Objective ·To investigate the value of the T-cell mitochondrial damage index (MDI) in risk prediction for acute exacerbation of chronic obstructive pulmonary disease (AECOPD). Methods ·A total of 100 patients with chronic obstructive pulmonary disease (COPD) who attended the Department of Respiratory Medicine, Jinhua Municipal Central Hospital, from May 2023 to December 2024 were enrolled. According to the post-bronchodilator percentage of predicted forced expiratory volume in 1 second (FEV₁%pred), they were divided into the moderate COPD group (50%≤FEV₁%pred<80%, n=50) and the severe COPD group (30%≤FEV₁%pred<50%, n=50). Additionally, 50 healthy individuals who underwent physical examinations during the same period were included as the healthy control group. Baseline characteristics of all subjects (including general clinical data and laboratory indicators) were collected. Flow cytometry was used to detect and compare the MDI of peripheral blood T cells and their subsets (CD3⁺, CD3⁺CD4⁺, and CD3⁺CD8⁺) in the three groups. The patients with COPD were followed up for 6 months, and were classified into the acute exacerbation (AE) group and the non-AE group according to whether AE occurred during follow-up. Differences in baseline characteristics between the two groups were compared. LASSO regression was used to screen baseline predictors of AECOPD. Univariable and multivariable Logistic regression models were used to establish a predictive model for AECOPD risk, and the performance of the model was validated using the receiver operating characteristic (ROC) curve, calibration curve, and five-fold cross-validation. Results ·The abnormal rates of MDI in T cells and their subsets were higher in patients with COPD than in the healthy control group. Among them, the abnormal rate of CD3⁺CD8⁺ T-cell MDI in the severe COPD group was significantly higher than that in the moderate COPD group and the healthy control group (both P<0.05). After 6 months of follow-up, 46 patients with COPD developed AE. Compared with the non-AE group, the patients in the AE group had lower FEV₁ levels (P<0.001), and higher COPD assessment tests (CAT) scores (P<0.001). Moreover, there were statistically significant differences in the categorical distribution of MDI injury grades in T cells and their subsets between the two groups (all P<0.001), with a higher proportion of mild-to-severe injury observed in the AE group. LASSO regression identified four baseline predictors: CD3⁺ T-cell MDI, CD3⁺CD8⁺ T-cell MDI, FEV₁, and CAT score. The area under the curve (AUC) of the AECOPD risk prediction model constructed based on these four variables was 0.848 (95%CI 0.772—0.923, P<0.001), with a sensitivity of 70.0% and a specificity of 81.0%. The calibration curve showed good model fit (P=0.136), and in the five-fold cross-validation, the accuracies of the training set and validation set was 0.80±0.03 and 0.82±0.06, respectively. Conclusion ·The AECOPD risk prediction model constructed based on CD3⁺ T-cell MDI, CD3⁺CD8⁺ T-cell MDI, FEV₁, and CAT score has good predictive value and may provide a reference for early clinical intervention.

Key words: chronic obstructive pulmonary disease (COPD), acute exacerbation (AE), T cell and its subset, mitochondrial damage index (MDI), prediction model

中图分类号:

R563.9

邓银灿, 陈晶, 王临英, 郭雪晶, 钱旭波, 朱丹. T细胞线粒体损伤指数在慢性阻塞性肺疾病急性加重风险预测中的价值[J]. 上海交通大学学报（医学版）, 2026, 46(5): 633-641.

Deng Yincan, Chen Jing, Wang Linying, Guo Xuejing, Qian Xubo, Zhu Dan. Value of T-cell mitochondrial damage index in predicting the risk of acute exacerbation of chronic obstructive pulmonary disease[J]. Journal of Shanghai Jiao Tong University (Medical Science), 2026, 46(5): 633-641.

图/表 7

参考文献 22

[1]	Wang C, Xu J Y, Yang L, et al. Prevalence and risk factors of chronic obstructive pulmonary disease in China (the China Pulmonary Health [CPH] study): a national cross-sectional study[J]. Lancet, 2018, 391(10131): 1706-1717.
[2]	张前豹, 施斌, 王小雨, 等. 无创呼吸机治疗COPD急性加重期老年患者发生抗生素相关腹泻的危险因素分析[J]. 重庆医学, 2024, 53(16): 2458-2461.
	Zhang Q B, Shi B, Wang X Y, et al. Analysis on risk factors of antibiotic related diarrhea occurrence in elderly patients with acute exacerbation of COPD treated by non-invasive ventilator[J]. Chongqing Medical Journal, 2024, 53(16): 2458-2461.
[3]	孔德昭, 王科雯, 罗文晔, 等. 慢性阻塞性肺疾病急性加重期诊疗指南评价与综合分析[J]. 中国中医基础医学杂志, 2023, 29(8): 1307-1317.
	Kong D Z, Wang K W, Luo W Y, et al. Evaluation and comprehensive analysis of diagnostic and treatment guidelines for acute exacerbation of chronic obstructive pulmonary disease[J]. Journal of Basic Chinese Medicine, 2023, 29(8): 1307-1317.
[4]	中华医学会呼吸病学分会慢性阻塞性肺疾病学组, 中国医师协会呼吸医师分会慢性阻塞性肺疾病工作委员会. 慢性阻塞性肺疾病诊治指南(2021年修订版)[J]. 中华结核和呼吸杂志, 2021, 44(3): 170-205.
	Chronic Obstructive Pulmonary Disease Group of Chinese Thoracic Society, Chronic Obstructive Pulmonary Disease Committee of Chinese Association of Chest Physician. Guidelines for the diagnosis and management of chronic obstructive pulmonary disease (revised version 2021)[J]. Chinese Journal of Tuberculosis and Respiratory Diseases, 2021, 44(3): 170-205.
[5]	GBD 2016 Causes of Death Collaborators. Global, regional, and national age-sex specific mortality for 264 causes of death, 1980-2016: a systematic analysis for the Global Burden of Disease Study 2016[J]. Lancet, 2017, 390(10100): 1151-1210.
[6]	Zhou M G, Wang H D, Zeng X Y, et al. Mortality, morbidity, and risk factors in China and its provinces, 1990-2017: a systematic analysis for the Global Burden of Disease Study 2017[J]. Lancet, 2019, 394(10204): 1145-1158.
[7]	吴玉兰, 陈周慰, 杨雪敏, 等. 慢性阻塞性肺疾病稳定期患者营养状况与CAT评分、肺功能的相关性及其风险预测Nomogram模型构建[J]. 临床肺科杂志, 2025, 30(6): 901-906.
	Wu Y L, Chen Z W, Yang X M, et al. Correlation between nutritional status and CAT score, pulmonary function, and risk prediction Nomogram model construction in stable COPD patients[J]. Journal of Clinical Pulmonary Medicine, 2025, 30(6): 901-906.
[8]	Soeroto A Y, Setiawan D, Asriputri N N, et al. Association between vitamin D levels and FEV1, number of exacerbations, and CAT score in stable COPD patients in Indonesia[J]. Int J Gen Med, 2021, 14: 7293-7297.
[9]	Phillips K M, Lavere P F, Hanania N A, et al. The emerging biomarkers in chronic obstructive pulmonary disease: a narrative review[J]. Diagnostics, 2025, 15(10): 1245.
[10]	Gonçalves I, Guimarães M J, van Zeller M, et al. Clinical and molecular markers in COPD[J]. Pulmonology, 2018, 24(4): 250-259.
[11]	Barnes P J. COPD 2020: new directions needed[J]. Am J Physiol Lung Cell Mol Physiol, 2020, 319(5): L884-L886.
[12]	林江涛. 呼吸内科学科进展报告[M]. 北京: 人民卫生出版社, 2014: 129-130.
	Lin J T. Progress report on respiratory medicine[M]. Beijing: People′s Medical Publishing House, 2014: 129-130.
[13]	Tan D B A, Fernandez S, Price P, et al. Impaired function of regulatory T-cells in patients with chronic obstructive pulmonary disease (COPD)[J]. Immunobiology, 2014, 219(12): 975-979.
[14]	Belk J A, Daniel B, Satpathy A T. Epigenetic regulation of T cell exhaustion[J]. Nat Immunol, 2022, 23(6): 848-860.
[15]	Franco F, Jaccard A, Romero P, et al. Metabolic and epigenetic regulation of T-cell exhaustion[J]. Nat Metab, 2020, 2(10): 1001-1012.
[16]	张怡, 李超, 张烈, 等. 慢阻肺患者运动负荷气道反应性与T细胞亚群的关系[J]. 现代生物医学进展, 2022, 22(24): 4710-4713, 4699.
	Zhang Y, Li C, Zhang L, et al. The relationship between exercise load airway responsiveness and T cell subsets in patients with chronic obstructive pulmonary disease[J]. Progress in Modern Biomedicine, 2022, 22(24): 4710-4713, 4699.
[17]	Geltink R I K, Kyle R L, Pearce E L. Unraveling the complex interplay between T cell metabolism and function[J]. Annu Rev Immunol, 2018, 36: 461-488.
[18]	Wang J, Yue H Y, Dong Y Z, et al. Effective compound combination of Bufei Yishen formula ameliorates PM_2.5-induced COPD by inhibiting mitochondrial oxidative stress through SIRT3-mediated FOXO3 deacetylation[J]. Phytomedicine, 2025, 140: 156568.
[19]	Wrench C L, Baker J R, Monkley S, et al. Small airway fibroblasts from patients with chronic obstructive pulmonary disease exhibit cellular senescence[J]. Am J Physiol Lung Cell Mol Physiol, 2024, 326(3): L266-L279.
[20]	Richter F C, Saliutina M, Hegazy A N, et al. Take my breath away-mitochondrial dysfunction drives CD8⁺ T cell exhaustion[J]. Genes Immun, 2024, 25(1): 4-6.
[21]	Park D W, Zmijewski J W. Mitochondrial dysfunction and immune cell metabolism in sepsis[J]. Infect Chemother, 2017, 49(1): 10-21.
[22]	Barberis M, Rojas López A. Metabolic imbalance driving immune cell phenotype switching in autoimmune disorders: tipping the balance of T- and B-cell interactions[J]. Clin Transl Med, 2024, 14(3): e1626. ［本文编辑］邢宇洋

Item	Control group (n=50)	Moderate COPD group (n=50)	Severe COPD group (n=50)	t/F/Z/χ² value	P value
Age/year	59.38±7.50	65.66±7.59^#	66.64±5.80^#	15.774	<0.001
Male/n(%)	40 (80.00)	44 (88.00)	41 (82.00)	1.248	0.536
Smoking/n(%)	8 (16.00)	33 (66.00)	40 (80.00)	41.451	<0.001
FEV₁/L	2.77±0.63	1.58±0.29^#	0.94±0.22^#*	243.105	<0.001
CAT score/score	‒	18±2	27±3	16.156	<0.001
IL-2/(pg·mL^-1)	3.15 (1.66, 4.28)	1.96 (1.11, 3.71)^#	2.01 (1.13, 2.81)^#	8.600	0.014
IL-4/(pg·mL^-1)	4.02 (2.80, 5.37)	1.27 (0.71, 1.82)^#	1.42 (0.90, 1.70)^#	67.170	<0.001
IL-6/(pg·mL^-1)	2.41 (1.39, 3.62)	1.89 (0.87, 5.27)	1.61 (1.14, 2.34)	2.671	0.263
IL-8/(pg·mL^-1)	2.30 (1.77, 3.87)	2.15 (0.03, 6.31)	1.48 (0.01, 3.34)	5.383	0.068
IL-10/(pg·mL^-1)	2.74 (1.46, 5.65)	1.45 (0.79, 2.30)^#	1.60 (1.42, 1.99)^#	19.710	<0.001
IL-17/(pg·mL^-1)	2.64 (1.66, 5.22)	1.96 (1.13, 3.21)	2.19 (1.62, 3.05)	5.044	0.080
TNF-α/(pg·mL^-1)	3.11 (1.85, 5.59)	1.15 (0.04, 1.77)^#	1.38 (0.42, 1.82)^#	52.177	<0.001
IFN-γ/(pg·mL^-1)	2.91 (1.70, 5.24)	2.45 (1.19, 3.88)	2.77 (1.61, 3.81)	3.722	0.156
CD3⁺ T absolute count/μL	1 031.50 (819.25, 1 309.00)	880.65 (643.48, 1 365.14)	913.09 (643.21, 1 128.30)	2.837	0.242
CD3⁺CD4⁺ T absolute count/μL	573.50 (448.25, 744.75)	566.68 (320.86, 824.50)	535.72 (373.98, 672.43)	0.736	0.692
CD3⁺CD8⁺ T absolute count/μL	363.50 (261.25, 460.00)	293.87 (194.05, 428.98)	285.51 (196.94, 454.26)	2.066	0.356
CD4⁺/CD8⁺ T ratio	1.64 (1.28, 2.19)	1.75 (1.29, 2.52)	1.67 (1.17, 2.34)	0.734	0.693
CD3⁺ T MDI/n(%)				50.267	<0.001
Normal	49 (98.00)	24 (48.00)^#	17 (34.00)^#
Mild damage	1 (2.00)	10 (20.00)	19 (38.00)
Severe damage	0 (0)	16 (32.00)	14 (28.00)
CD3⁺CD4⁺ T MDI/n(%)				55.779	<0.001
Normal	49 (98.00)	22 (44.00)^#	14 (28.00)^#
Mild damage	1 (2.00)	24 (48.00)	28 (56.00)
Severe damage	0 (0)	4 (8.00)	8 (16.00)
CD3⁺CD8⁺ T MDI/n(%)				63.779	<0.001
Normal	49 (98.00)	25 (50.00)^#	10 (20.00)^#
Mild damage	1 (2.00)	18 (36.00)	25 (50.00)
Severe damage	0 (0)	7 (14.00)	15 (30.00)^*

Item	Control group (n=50)	Moderate COPD group (n=50)	Severe COPD group (n=50)	t/F/Z/χ² value	P value
Age/year	59.38±7.50	65.66±7.59^#	66.64±5.80^#	15.774	<0.001
Male/n(%)	40 (80.00)	44 (88.00)	41 (82.00)	1.248	0.536
Smoking/n(%)	8 (16.00)	33 (66.00)	40 (80.00)	41.451	<0.001
FEV₁/L	2.77±0.63	1.58±0.29^#	0.94±0.22^#*	243.105	<0.001
CAT score/score	‒	18±2	27±3	16.156	<0.001
IL-2/(pg·mL^-1)	3.15 (1.66, 4.28)	1.96 (1.11, 3.71)^#	2.01 (1.13, 2.81)^#	8.600	0.014
IL-4/(pg·mL^-1)	4.02 (2.80, 5.37)	1.27 (0.71, 1.82)^#	1.42 (0.90, 1.70)^#	67.170	<0.001
IL-6/(pg·mL^-1)	2.41 (1.39, 3.62)	1.89 (0.87, 5.27)	1.61 (1.14, 2.34)	2.671	0.263
IL-8/(pg·mL^-1)	2.30 (1.77, 3.87)	2.15 (0.03, 6.31)	1.48 (0.01, 3.34)	5.383	0.068
IL-10/(pg·mL^-1)	2.74 (1.46, 5.65)	1.45 (0.79, 2.30)^#	1.60 (1.42, 1.99)^#	19.710	<0.001
IL-17/(pg·mL^-1)	2.64 (1.66, 5.22)	1.96 (1.13, 3.21)	2.19 (1.62, 3.05)	5.044	0.080
TNF-α/(pg·mL^-1)	3.11 (1.85, 5.59)	1.15 (0.04, 1.77)^#	1.38 (0.42, 1.82)^#	52.177	<0.001
IFN-γ/(pg·mL^-1)	2.91 (1.70, 5.24)	2.45 (1.19, 3.88)	2.77 (1.61, 3.81)	3.722	0.156
CD3⁺ T absolute count/μL	1 031.50 (819.25, 1 309.00)	880.65 (643.48, 1 365.14)	913.09 (643.21, 1 128.30)	2.837	0.242
CD3⁺CD4⁺ T absolute count/μL	573.50 (448.25, 744.75)	566.68 (320.86, 824.50)	535.72 (373.98, 672.43)	0.736	0.692
CD3⁺CD8⁺ T absolute count/μL	363.50 (261.25, 460.00)	293.87 (194.05, 428.98)	285.51 (196.94, 454.26)	2.066	0.356
CD4⁺/CD8⁺ T ratio	1.64 (1.28, 2.19)	1.75 (1.29, 2.52)	1.67 (1.17, 2.34)	0.734	0.693
CD3⁺ T MDI/n(%)				50.267	<0.001
Normal	49 (98.00)	24 (48.00)^#	17 (34.00)^#
Mild damage	1 (2.00)	10 (20.00)	19 (38.00)
Severe damage	0 (0)	16 (32.00)	14 (28.00)
CD3⁺CD4⁺ T MDI/n(%)				55.779	<0.001
Normal	49 (98.00)	22 (44.00)^#	14 (28.00)^#
Mild damage	1 (2.00)	24 (48.00)	28 (56.00)
Severe damage	0 (0)	4 (8.00)	8 (16.00)
CD3⁺CD8⁺ T MDI/n(%)				63.779	<0.001
Normal	49 (98.00)	25 (50.00)^#	10 (20.00)^#
Mild damage	1 (2.00)	18 (36.00)	25 (50.00)
Severe damage	0 (0)	7 (14.00)	15 (30.00)^*

Item	AE group (n=46)	Non-AE group (n=54)	t/Z/χ² value	P value
Age/year	67.13±5.70	65.31±7.47	1.348	0.181
Male/n(%)	42 (91.30)	43 (79.63)	1.819	0.177
Smoking/n(%)	35 (76.08)	36 (66.67)	1.071	0.301
COPD/n(%)			4.026	0.045
Moderate	18 (39.13)	32 (59.26)
Severe	28 (60.87)	22 (40.74)
Baseline FEV₁/L	1.10±0.37	1.40±0.38	3.908	<0.001
ΔFEV₁/L	-0.03 (-0.12, 0.09)	0.12 (0.06, 0.26)	22.311	<0.001
Baseline CAT score/score	25±5	21±4	4.135	<0.001
ΔCAT score/score	7 (4, 9)	9 (7, 11)	10.099	0.001
IL-2/(pg·mL^-1)	2.08 (1.41, 3.09)	1.93 (1.01, 2.95)	0.741	0.389
IL-4/(pg·mL^-1)	1.45 (0.95, 1.81)	1.16 (0.84, 1.74)	1.971	0.160
IL-10/(pg·mL^-1)	1.55 (1.21, 2.03)	1.56 (1.07, 2.30)	0.153	0.696
TNF-α/(pg·mL^-1)	1.45 (0.77, 1.96)	1.11 (0.04, 1.50)	2.914	0.088
CD3⁺ T absolute count/μL	932.26 (609.06, 1 229.28)	880.65 (594.97, 1 114.72)	0.553	0.580
CD3⁺CD4⁺ T absolute count/μL	528.69 (372.02, 796.94)	568.74 (393.69, 727.02)	0.367	0.714
CD3⁺CD8⁺ T absolute count/μL	323.83 (207.97, 447.78)	265.47 (233.07, 421.11)	0.844	0.399
CD4⁺/CD8⁺ T ratio	1.68 (1.43, 2.33)	1.95 (1.49, 2.82)	0.470	0.638
CD3⁺ T MDI/n(%)			15.047	<0.001
Normal	10 (21.74)	31 (57.41)
Mild damage	15 (32.61)	14 (25.92)
Severe damage	21 (45.65)	9 (16.67)
CD3⁺CD4⁺ T MDI/n(%)			15.291	<0.001
Normal	10 (21.74)	26 (48.15)
Mild damage	26 (56.52)	27 (50.00)
Severe damage	10 (21.74)	1 (1.85)
CD3⁺CD8⁺ T MDI/n(%)			25.390	<0.001
Normal	7 (15.22)	30 (55.56)
Mild damage	20 (43.48)	20 (37.03)
Severe damage	19 (41.30)	4 (7.41)

Item	AE group (n=46)	Non-AE group (n=54)	t/Z/χ² value	P value
Age/year	67.13±5.70	65.31±7.47	1.348	0.181
Male/n(%)	42 (91.30)	43 (79.63)	1.819	0.177
Smoking/n(%)	35 (76.08)	36 (66.67)	1.071	0.301
COPD/n(%)			4.026	0.045
Moderate	18 (39.13)	32 (59.26)
Severe	28 (60.87)	22 (40.74)
Baseline FEV₁/L	1.10±0.37	1.40±0.38	3.908	<0.001
ΔFEV₁/L	-0.03 (-0.12, 0.09)	0.12 (0.06, 0.26)	22.311	<0.001
Baseline CAT score/score	25±5	21±4	4.135	<0.001
ΔCAT score/score	7 (4, 9)	9 (7, 11)	10.099	0.001
IL-2/(pg·mL^-1)	2.08 (1.41, 3.09)	1.93 (1.01, 2.95)	0.741	0.389
IL-4/(pg·mL^-1)	1.45 (0.95, 1.81)	1.16 (0.84, 1.74)	1.971	0.160
IL-10/(pg·mL^-1)	1.55 (1.21, 2.03)	1.56 (1.07, 2.30)	0.153	0.696
TNF-α/(pg·mL^-1)	1.45 (0.77, 1.96)	1.11 (0.04, 1.50)	2.914	0.088
CD3⁺ T absolute count/μL	932.26 (609.06, 1 229.28)	880.65 (594.97, 1 114.72)	0.553	0.580
CD3⁺CD4⁺ T absolute count/μL	528.69 (372.02, 796.94)	568.74 (393.69, 727.02)	0.367	0.714
CD3⁺CD8⁺ T absolute count/μL	323.83 (207.97, 447.78)	265.47 (233.07, 421.11)	0.844	0.399
CD4⁺/CD8⁺ T ratio	1.68 (1.43, 2.33)	1.95 (1.49, 2.82)	0.470	0.638
CD3⁺ T MDI/n(%)			15.047	<0.001
Normal	10 (21.74)	31 (57.41)
Mild damage	15 (32.61)	14 (25.92)
Severe damage	21 (45.65)	9 (16.67)
CD3⁺CD4⁺ T MDI/n(%)			15.291	<0.001
Normal	10 (21.74)	26 (48.15)
Mild damage	26 (56.52)	27 (50.00)
Severe damage	10 (21.74)	1 (1.85)
CD3⁺CD8⁺ T MDI/n(%)			25.390	<0.001
Normal	7 (15.22)	30 (55.56)
Mild damage	20 (43.48)	20 (37.03)
Severe damage	19 (41.30)	4 (7.41)

Variable	Univariate Logistic regression analysis		Multivariate Logistic regression analysis
Variable	OR (95%CI)	P value	OR (95%CI)	P value
FEV₁	0.13 (0.04‒0.40)	<0.001	0.24 (0.02‒0.64)	0.035
CAT score	1.18 (1.08‒1.29)	<0.001	1.12 (1.06‒1.28)	0.025
CD3⁺ T MDI
Normal	1.00	‒	1.00	‒
Mild damage	3.32 (1.20‒9.20)	0.021	1.93 (1.34‒6.35)	0.041
Severe damage	7.23 (2.51‒20.82)	<0.001	4.41 (1.21‒16.08)	0.024
CD3⁺CD8⁺ T MDI
Normal	1.00	‒	1.00	‒
Mild damage	4.29 (1.53‒12.01)	0.006	3.18 (1.04‒9.76)	0.043
Severe damage	20.36 (5.25‒79.02)	<0.001	8.83 (2.00‒38.96)	0.004