Evaluation of three molecular methods for the detection of drug resistance of Mycobacterium tuberculosis

doi:10.3969/j.issn.1000-6621.2019.04.014

Abstract

Abstract:

Objective To analyze the value of GeneXpert MTB/RIF (GeneXpert), gene chip and MeltPro TB in the detection of drug resistance of Mycobacterium tuberculosis (MTB). Methods A total of 774 cases of pulmonary TB with positive culture of MTB in sputum from January 2017 to January 2018 in Xi’an Chest Hospital were retrospectively analyzed. The results of phenotypic drug sensitivity test, GeneXpert detection, gene chip detection and MeltPro TB detection were collected. Based on phenotypic drug susceptibility criteria, the efficacy of the three molecular biology methods to detect drug resistance of MTB was analyzed. The accuracy of the three molecular biology methods to detect rifampicin (RFP) resistance was compared according to the area under receiver operating characteristic (ROC) curve. Results With phenotypic drug susceptibility as standard, the sensitivity and specificity were 95.6% (87/91) and 93.6% (249/266) for GeneXpert in detecting RFP resistance; and for gene chip in detecting RFP and isoniazid (INH) resistance, the sensitivity were 92.8% (90/97) and 78.8% (130/165), and the specificity were 96.8% (610/630) and 99.1% (557/562). For MeltPro TB in detecting RFP, INH, levofloxacin (Lfx) and amikacin (Am), the sensitivity were 94.4% (51/54), 82.4% (61/74), 87.1% (27/31) and 75.0% (6/8), and the specificity were 93.9% (138/147), 95.3% (121/127), 98.8% (164/166) and 99.5% (188/189). The area under the ROC curve of GeneXpert in detecting RFP resistance was 0.95±0.02; the area under the ROC curve of gene chip and MeltPro TB in detecting RFP resistance were 0.93±0.03. Conclusion Detection of MTB resistance to the first-line anti-tuberculosis drugs by GeneXpert, gene chip and MeltPro TB was reliable, and detection of MTB resistance to Lfx and Am by MeltPro TB was reliable.

Key words: Tuberculosis, multidrug-resistant, Microbial sensitivity tests, Molecular probe techniques, Nucleic acid amplification techniques, Evaluation studies

Jin-bao MA,Han YANG,Fei REN,Li-yun DANG. Evaluation of three molecular methods for the detection of drug resistance of Mycobacterium tuberculosis[J]. Chinese Journal of Antituberculosis, 2019, 41(4): 440-446. doi: 10.3969/j.issn.1000-6621.2019.04.014

Figures/Tables 5

References 22

[1]	姜世闻, 胡屹, 张慧 , 等. 耐药结核病的流行简况. 中国防痨杂志, 2010,32(6):57-59.
[2]	郭苏珊, 林健雄, 郭金镇 , 等. BACTEC MGIT960用于结核分枝杆菌直接药敏试验的研究. 中国病原生物学杂志, 2014,9(10):946-952.
[3]	World Health Organization. Treatment guidelines for drug-resistant tuberculosis, 2016 update. Geneva: World Health Organization, 2016. URL pmid: 27748093
[4]	中华医学会结核病学分会临床检验专业委员会. 结核病病原学分子诊断专家共识. 中华结核和呼吸杂志, 2018,41(9):688-695.
[5]	中华人民共和国国家卫生和计划生育委员会.WS 288—2017肺结核诊断. 2017-11-09.
[6]	王黎霞 . 结核分枝杆菌药物敏感性试验标准化操作程序及质量保证手册. 北京: 人民卫生出版社, 2013.
[7]	Blakemore R, Story E, Helb D , et al. Evaluation of the analytical performance of the Xpert MTB/RIF assay. J Clin Microbiol, 2010,48(7):2495-2501. doi: 10.1128/JCM.00128-10 URL
[8]	赵玉玲, 石大伟, 贾琼 , 等. 河南省耐多药结核分枝杆菌利福平耐药相关基因rpoB突变分析. 中华结核和呼吸杂志, 2011,34(5):389-390. doi: 10.3760/cma.j.issn.1001-0939.2011.05.023 URL
[9]	牛海军, 王歌, 李明虎 . GeneXpert MTB/RIF 检测在肺结核诊断中的价值评估. 中国防痨杂志, 2017,39(8):829-832. doi: 10.3969/j.issn.1000-6621.2017.08.008 URL
[10]	路丽苹, 陆斌, 刘梅 , 等. GeneXpert MTB/RIF技术在结核病早诊断中的应用. 中国防痨杂志, 2017,39(1):71-75. doi: 10.3969/j.issn.1000-6621.2017.01.016 URL
[11]	中国防痨协会结核病临床专业委员会. 结核病临床诊治进展年度报告(2014年)(第一部分结核病临床诊断). 中国防痨杂志, 2015,37(6):549-582. doi: 10.3969/j.issn.1000-6621.2015.06.001 URL
[12]	唐神结, 许绍发, 李亮 . 耐药结核病学.北京: 人民卫生出版社, 2014: 118-125.
[13]	World Health Organization . Global tuberculosis report 2018. Geneva:World Health Organization, 2018.
[14]	World Health Organization. Rapid communication: key changes to treatment of multidrug- and rifampicin-resistant tuberculosi (MDR/RR-TB). Geneva: World Health Organization, 2018.
[15]	唐神结 . 氟喹诺酮类药物在耐药结核病中应用的再评价. 中华结核和呼吸杂志, 2014,37(10):727-729. doi: 10.3760/cma.j.issn.1001-0939.2014.10.004 URL
[16]	World Health Organization. The use of molecular line probe assays for the detection of resistance to second-line anti-tuberculosis drugs: policy guidance. Geneva: World Health Organization, 2016.
[17]	Pang Y, Dong H, Tan Y , et al. Rapid diagnosis of MDR and XDR tuberculosis with the MeltPro TB assay in China. Sci Rep, 2016,6:25330. doi: 10.1038/srep25330 URL pmid: 4858717
[18]	林明冠, 吴元东, 朱中元 , 等. 基因芯片技术在结核分枝杆菌耐药检测中的效果分析. 中国防痨杂志, 2018,40(1):58-62. doi: 10.3969/j.issn.1000-6621.2018.01.014 URL
[19]	Zhu L, Liu Q, Martinez L , et al. Diagnostic value of GeneChip for detection of resistant Mycobacterium tuberculosis in patients with differing treatment histories. J Clin Microbiol, 2015,53(1):131-135. doi: 10.1128/JCM.02283-14 URL pmid: 4290911
[20]	胡春梅, 郭晶, 严虹 , 等. 荧光定量PCR探针熔解曲线法在结核分枝杆菌耐药基因检测中的应用. 中国防痨杂志, 2016,38(1):38-41. doi: 10.3969/j.issn.1000-6621.2016.01.009 URL
[21]	王晓艳, 王云超, 王珍 , 等. 探针溶解曲线法检测痰结核杆菌耐异烟肼突变. 中国医学工程, 2015,11(23):68-69.
[22]	Zhang MJ, Ren WZ, Sun XJ , et al. GeneChip analysis of resistant Mycobacterium tuberculosis with previously treated tuberculosis in Changchun. BMC Infect Dis, 2018,18(1):234. doi: 10.1186/s12879-018-3131-8 URL

GeneXpert 检测	表型药敏试验		合计 (例)	敏感度 [%(95%CI值)]	特异度 [%(95%CI值)]	阳性预测值 [%(95%CI值)]	阴性预测值 [%(95%CI值)]	Kappa值
GeneXpert 检测	耐药(例)	敏感(例)	合计 (例)	敏感度 [%(95%CI值)]	特异度 [%(95%CI值)]	阳性预测值 [%(95%CI值)]	阴性预测值 [%(95%CI值)]	Kappa值
全部患者				95.6 (88.5~98.6)	93.6 (89.8~96.1)	83.7 (74.8~89.9)	98.4 (96.7~99.5)	0.85
耐药(例)	87	17	104
敏感(例)	4	249	253
合计(例)	91	266	357
初治患者				87.9 (70.9~96.0)	93.9 (89.8~96.5)	67.4 (51.3~80.4)	98.1 (95.1~99.4)	0.72
耐药(例)	29	14	43
敏感(例)	4	217	221
合计(例)	33	231	264
复治患者				100.0 (92.3~100.0)	91.4 (75.8~97.8)	95.1 (85.4~98.7)	100.0 (86.6~100.0)	0.93
耐药(例)	58	3	61
敏感(例)	0	32	32
合计(例)	58	35	93

基因芯片检测	表型药敏试验		合计 (例)	敏感度 [%(95%CI值)]	特异度 [%(95%CI值)]	阳性预测值 [%(95%CI值)]	阴性预测值 [%(95%CI值)]	Kappa值
基因芯片检测	耐药(例)	敏感(例)	合计 (例)	敏感度 [%(95%CI值)]	特异度 [%(95%CI值)]	阳性预测值 [%(95%CI值)]	阴性预测值 [%(95%CI值)]	Kappa值
全部患者				92.8 (85.2~96.8)	96.8 (95.1~98.0)	81.8 (73.1~88.3)	98.9 (97.6~99.5)	0.85
耐药(例)	90	20	110
敏感(例)	7	610	617
合计(例)	97	630	727
初治患者				91.1 (77.9~97.1)	97.2 (95.4~98.3)	71.9 (58.3~82.6)	99.3 (98.0~99.8)	0.79
耐药(例)	41	16	57
敏感(例)	4	555	559
合计(例)	45	571	616
复治患者				94.2 (83.1~98.5)	93.3 (82.7~97.8)	92.5 (80.9~97.6)	94.8 (84.7~98.7)	0.87
耐药(例)	49	4	53
敏感(例)	3	55	58
合计(例)	52	49	111

基因芯片检测	表型药敏试验		合计 (例)	敏感度 [%(95%CI值)]	特异度 [%(95%CI值)]	阳性预测值 [%(95%CI值)]	阴性预测值 [%(95%CI值)]	Kappa值
基因芯片检测	耐药(例)	敏感(例)	合计 (例)	敏感度 [%(95%CI值)]	特异度 [%(95%CI值)]	阳性预测值 [%(95%CI值)]	阴性预测值 [%(95%CI值)]	Kappa值
全部患者				78.8 (71.6~84.6)	99.1 (97.8~99.7)	96.3 (91.1~98.6)	94.1 (91.7~95.8)	0.83
耐药(例)	130	5	135
敏感(例)	35	557	592
合计(例)	165	562	727
初治患者				78.8 (69.5~86.0)	99.4 (98.1~99.8)	96.5 (89.3~99.1)	95.9 (93.7~97.3)	0.84
耐药(例)	82	3	85
敏感(例)	22	509	531
合计(例)	104	512	616
复治患者				78.7 (66~87.7)	96.0 (85.1~99.3)	96.0 (85.1~99.3)	78.7 (66.0~87.7)	0.73
耐药(例)	48	2	50
敏感(例)	13	48	61
合计(例)	61	50	111

PCR熔解曲线法	表型药敏试验		合计 (例)	敏感度 [%(95%CI值)]	特异度 [%(95%CI值)]	阳性预测值 [%(95%CI值)]	阴性预测值 [%(95%CI值)]	Kappa值
PCR熔解曲线法	耐药(例)	敏感(例)	合计 (例)	敏感度 [%(95%CI值)]	特异度 [%(95%CI值)]	阳性预测值 [%(95%CI值)]	阴性预测值 [%(95%CI值)]	Kappa值
利福平				94.4 (83.6~98.6)	93.9 (88.4~97.0)	85.0 (72.9~92.5)	97.9 (93.4~99.4)	0.85
耐药(例)	51	9	60
敏感(例)	3	138	141
合计(例)	54	147	201
异烟肼				82.4 (71.5~90.0)	95.3 (89.6~98.1)	91.0 (80.9~96.3)	90.3 (83.7~94.5)	0.79
耐药(例)	61	6	67
敏感(例)	13	121	134
合计(例)	74	127	201
左氧氟沙星				87.1 (69.2~95.8)	98.8 (95.3~99.8)	93.1 (75.8~98.8)	97.6 (93.6~99.2)	0.88
耐药(例)	27	2	29
敏感(例)	4	164	168
合计(例)	31	166	197
莫西沙星				87.0 (65.3~96.6)	94.8 (90.1~97.5)	69.0 (49.0~84.0)	98.2 (94.4~99.5)	0.70
耐药(例)	20	9	29
敏感(例)	3	165	168
合计(例)	23	174	197
阿米卡星				75.0 (35.6~95.5)	99.5 (96.6~99.9)	85.7 (42.0~99.2)	98.9 (95.8~99.8)	0.79
耐药(例)	6	1	7
敏感(例)	2	188	190
合计(例)	8	189	197
卷曲霉素				71.4 (30.3~94.9)	98.9 (95.8~99.8)	71.4 (30.3~94.9)	98.9 (95.8~99.8)	0.70
耐药(例)	5	2	7
敏感(例)	2	188	190
合计(例)	7	190	197