Objective   To evaluate the feasibility and security of Liquid-based Peng’s interlayer vessel technique used for diagnosis of tuberculosis in primary level laboratory.  Methods   In two primary level laboratories of Taixing and Jiangyin Center for Disease Control and Prevention, 754 sputum samples from 385 suspect TB patients were tested simultaneously by two methods: sputum smear and acid-fast staining microscopy examination (direct AFB) and Liquid-based Peng’s interlayer vessel technique. According to clinical diagnosis, the positive rates of the two methods for the sputum samples with different traits or collected at different times were compared by Fisher probabilities in 2×2 table. 12 bacterium-positive sputa were treated with different proportion of digestion liquid, different digestion time, and heating or not. Then the digestive sputa were inoculated at Roche medium at  37℃. The security of the liquid-based method was evaluated by preparing the growth of mycobacteria.   Results   The positive rates of Liquid-based Peng’s interlayer vessel technique and the direct AFB were 23.5% (177/754) and 16.8% (127/754) respectively, which had significant difference ( χ 2 =48.02, P =0.000). The positive rates of these two methods in purulent sputa were 47.5% (94/198) and 40.9% (81/198), which had significant differences ( P P Conclusion   Liquid-based Peng’s interlayer vessel technique can improve the positive detection rate in suspect TB patients. It is simple, safe, and suitable for the application at primary laboratory for the diagnosis of tuberculosis and convenient for quality control.

目的   评价液基夹层杯法在基层实验室用于结核病诊断的可行性及安全性。 方法   在泰兴、江阴CDC两个县级结核病实验室,对385例疑似肺结核患者的754份痰标本同时采用液基夹层杯法与直接痰涂片法进行镜检,结合临床诊断，比较2种方法的阳性检出率。采用四格表Fisher确切概率法进行比较，并进一步比较两种方法对不同性状、不同时间收集的痰标本的阳性率。选取12份阳性痰标本,设定不同的痰消化液比例,不同的消化时间,加热与否处理痰标本，将处理后的消化液接种到罗氏培养基，37℃培养，通过比较生长情况评价液基法的安全性。 结果   液基夹层杯法、直接痰涂片法对754份痰标本的阳性检出率分别为23.5%(177/754)、16.8%(127/754)，检测结果差异有统计学意义（ χ 2 =48.02， P =0.000）；比较不同性状、不同时间采集的痰标本采用两种方法的检测结果，得出脓痰直涂法阳性检出率为40.9%（81/198），液基法阳性检出率为47.5%（94/198），两种方法对于脓痰的检测结果差异有统计学意义（Fisher确切概率法， P 确切概率 ＜0.001）；非脓性痰（包括血痰、黏液痰、口水痰）直涂法阳性检出率为4.7%（9/193），液基法阳性检出率为13.5%(26/193)，差异有统计学意义（ P 确切概率 值均＜0.001）。安全性试验发现经消化液处理的痰标本，培养8周后未见结核分枝杆菌生长。 结论   液基夹层杯法提高了疑似肺结核患者细菌学检查的阳性检出率,且操作简单、安全性高，适合在基层实验室应用并便于开展质量控制,是一种很有推广价值的检测方法。

Objective To compare the using value of the automatic BACTEC MGIT960 system with Lowenstein-Jensen (L-J) culture method in culture, identification and drug susceptibility test of Mycobacterium. Methods A total of 435 cases (including 319 male patients and 116 female patients with average age 52.3 years) were detected by sputum smear for acid-fast staining, L-J culture and MGIT960 liquid culture, respectively. The culture-positive samples were further performed drug susceptibility test on isoniazid (INH), rifampicin (RIF), streptomycin (STR) and ethambutol (EMB) using the absolute concentration method on L-J medium and MGIT960 method, respectively. Their culture-positive rate, culture time and coincidence rate were analyzed.  Results The positive coincidence rate of MGIT960 and L-J culture with sputum smear were 90.8% (187/206) and 89.3% (184/206), respectively. The coincidence rate of MGIT960 and L-J culture was 92.8% (219/236). Their average culture times were 9.5 days and 31.5 days, respectively. The coincidence rate of INH、RIF、STR and EMB drug susceptibility test using MGIT960 and L-J culture method was 92.7%, 95.9%, 91.8% and 97.3%, respectively.  Conclusions The sensitivity and specificity of automatic MGIT960 are higher than L-J method. The report time of MGIT960 reduced to 9.5 days. The automatic MGIT960 is ideal for the rapid mycobacterial culture, identification and drug susceptibility test.

目的 比对全自动Bactec MGIT960（MGIT960）系统与改良罗氏（L-J）培养 方法 2者在结核分枝杆菌培养、鉴定及药敏试验中的优越性。 方法 按国家疾病预防控制中心（CDC）-碧迪项目辽宁工作方案要求和国际标准WS288-2009《肺结核诊断标准》纳入临床病例。本实验共纳入病例435例，其中男性319例，女性116例，平均年龄52.3岁（10～86岁）。上述每份病例标本均分别进行痰液涂片抗酸染色（萋-尼法）镜检，L-J固体培养和MGIT960系统液体培养。培养阳性者分别进行一线药物异烟肼（INH）、利福平（RIF）、链霉素（STR）和乙胺丁醇（EMB）的L-J法药敏试验（比例法）和MGIT960药敏试验；分别计算出抗酸染色镜检与MGIT960和L-J培养的符合率； L-J培养和MGIT960培养阳性检出率；L-J和MGIT960两种培养方法药敏试验所需的培养时间及符合率等。 结果MGIT960培养法和L-J培养法与痰涂片镜检的阳性符合率分别为90.8%（187/206）和89.3%（184/206）；MGIT960与L-J结果符合率92.8%（219/236）；MGIT960和L-J平均培养时间分别为9.5 d和31.5 d；MGIT960和L-J一线药物INH、RIF、STR和EMB敏感性试验的符合率分别为92.7%、95.9%、91.8%和97.3%。 结论 全自动MGIT960系统培养方法的敏感度和特异度高于L-J培养法，报告时间平均缩短至9.5-d，该系统是目前最为理想的快速结核分枝杆菌培养、鉴定和药敏试验检测系统。

Objective To systematically review the diagnostic accuracy of gene chip in identifying rifampicin (RFP) resistance Mycobacterium tuberculosis (Mtb)，and to provide the evidence for the clinical use of gene chip in detection of RFP-resistance Mtb.  Methods We electronically searched The Cochrane Library (Issue 1, 2014), PubMed, EMbase, Wanfang Data, CNKI from the date that the database was established to February 2014. In order to find any publications which are related to detection of RFP-resistance Mtb by using gene chip, the search key terms in English and Chinese were designed as follows: tuberculosis, tuberculosis and drug resistance, tuberculosis and drug resistance and gene chip, tuberculosis and drug resistance and gene chip and rifampicin. Two reviewers independently screened literatures according to the inclusion and exclusion criteria, extracted data and assessed the quality of methodology of the included studies according to the QUADAS items. The Meta-DiSc software (version 1.4) was used to conduct pooling on sensitivity (SEN), specificity (SPE), positive likelihood ratio (+LR), and negative likelihood ratio (-LR). Heterogeneity test was performed and the summary receiver operating characteristic (SROC) curve was drawn and area under curve (AUC) was calculated.  Results A total of 223 publications were found. Finally, 22 articles (including one article from multicenter study) and 25 studies were included into the review，which involved 4879 Mtb isolates (including 1314 RFP-resistant strains, 3565 susceptible strains and the standard strain H37Rv). The results of Meta-analysis showed that, compared with the traditional drug sensitivity test, SEN of gene chip in detection of RFP-resistance Mtb was 0.89 (95% CI (0.87－0.91)), SPE was 0.98 (95% CI (0.97－0.98)), +LR was 30.89 (95% CI (22.15－43.06)), -LR was 0.10 (95% CI (0.07－0.15)), diagnostic OR was 354.06 (95% CI (212.09－591.07)), AUC was 0.9833.  Conclusion Gene chip has a good value in detection of RFP-resistance Mtb. If the traditional drug sensitivity test is regarded as a gold standard, gene chip has high sensitivity, high specificity and high diagnostic OR value. It can be used as a clinical supplementary method of drug sensitivity test.

目的 对基因芯片方法在检测耐利福平结核分枝杆菌的准确性进行系统评价，为基因芯片技术应用于临床检测耐药结核分枝杆菌的必要性提供证据。 方法 计算机检索 The Cochrane Library （2014年第1期）、PubMed、EMbase、Wanfang Data、CNKI数据库，检索时限均为建库至2014年2月。由2位研究者根据纳入与排除标准独立筛选文献、提取资料，采用评价诊断准确性质量研究（quality assessment of diagnostic accuracy studies，QUADAS）条目评价纳入研究论文的方法学质量，检出相关文献223篇,按照标准纳入22篇文献。采用Meta-DiSc 1.4软件对其敏感度（SEN）、特异度（SPE）、阳性似然比（+LR）、阴性似然比（－LR）、诊断比值比（DOR）进行异质性检验和合并分析，绘制汇总受试者工作特征（SROC）曲线，计算曲线下面积（AUC）。 结果 共纳入22篇文献(包括一篇多中心研究文献)，25个研究，包括4879株经传统药敏试验鉴定结核分枝杆菌菌株，其中耐RFP菌株（1314株）、敏感菌株（3565株）。基因芯片技术检测Mtb对RFP耐药的SEN 合并 为0.89［95% CI (0.87～0.91)］、SPE 合并 为0.98［95% CI (0.97～0.98)］、+LR为30.89［95% CI (22.15～43.06)］、－LR为0.10［95% CI (0.07～0.15)］、DOR 合并 为354.06［95% CI (212.09～591.07)］、AUC为0.9833。 结论 基因芯片方法具有较好的诊断价值；以传统药敏法为金标准，基因芯片方法特异度、敏感度，诊断 OR 值均较高，说明其在Mtb对RFP耐药检出率较高，可作为临床结核分枝杆菌耐药检测的辅助手段。

一、基因芯片技术概述基因芯片(又称DNA芯片)始创于90年代初,首先由美国Affymetrix公司的Fodor博士提出并开始基因芯片技术的研究.它是目前分子生物学最前沿的方法,是物理学、微电子学与生命科学交叉综合的高新技术,是近年来发展起来的进行大规模遗传多态性检测的新方法[1].基因芯片以其基片的不同分为无机基片和有机合成物基片,前者包括半导体硅片和玻璃片,其上的探针主要以原位聚合的方法合成;而后者的探针是预先合成后通过特殊的微量点样装置或仅器滴加到基片上去[2].基因芯片的基本原理是将多种探针固定在玻璃等基片上,然后与待测样本的DNA或RNA进行杂交,通过检测每个探针分子的杂交信号强度进而获取样品分子的数量和序列信息.由于该法同时将大量探针固定于支持物上,所以一次可以对样品大量序列进行检测和分析,从而解决了传统核酸印迹技术的操作繁杂、自动化程度低、操作序列数量少、检测效率低等不足[3].

Objective Study on gene chip technique in detection of drug resistance of Mycobacterium tuberculosis and its clinical value.   Methods (1) One hundred and thirty-seven strain identification of Mycobacterium clinical isolates from Changsha Central Hospital by gene chip and Spoligotyping technology from 2011-01 to 2012-12. (2)Drug-susceptibility test of rifampin and isoniazid to Mycobacterium tuberculosis clinical isolates by absolute concentration method. (3) Drug-susceptibility test of rifampin and isoniazid to Mycobacterium tuberculosis clinical isolates by DNA chips for rpoB , KatG , inhA . Rifampin resistant is one or a plurality of site mutation of rpoB gene, isoniazid resistance is one or a plurality of site mutation of KatG gene and inhA gene. Compared Spoligotyping with gene chip method of species identification by Kappa test, Compared the absolute concentration method with gene chip method of drug sensitivity by χ 2 test and Kappa test.   Results （1）For 45 rifampin sensitive (under 50 μg/ml) and isoniazid sensitive(under 1 μg/ml) isolates tested by absolute concentration method, 42 are wild type for rpoB gene with the sensitivity of 93.3%(1 for 511T-C mutations, 1 for 531C-T mutations, 1 for 516A-T mutations), 45 are wild type for KatG gene with the sensitivity of 100.0% and 43 are wild type for inhA gene with the sensitivity of 95.6%(2 for15C-T mutation) by DNA chip method, respectively. （2）For 18 rifampin-resistant (50－250 μg/ml) isolates tested by absolute concentration method, 16 has mutation in rpoB gene with the sensitivity of 88.9% by DNA chip method. Those mutations happened at 531, 516, 526, 511 amino acid sites of rpoB . For 19 rifampin-resistant (above 250 μg/ml) isolates tested by absolute concentration method, 19 has mutation in rpoB gene with the sensitivity of 100% by DNA chip method. Those mutations happened at 531, 516, 526, 511 amino acid sites of rpoB .（3）Compare the DNA chips with the absolute concentration method, for 13 isoniazid-resistance(1－10 μg/ml) isolates, 12 has mutation in KatG gene with the sensitivity of 92.3% by DNA chip method, which mainly happened at 315 site. No mutation related to the drug resistance was found in inhA gene. Compare the DNA chips with the absolute Concentration method, For 9 isoniazid-resistance(above 10 μg/ml) isolates, 9 has mutation in KatG gene with the sensitivity of 100% by DNA chip method, which mainly happened at 315 site. No mutation related to the drug resistance was found in inhA gene. （4）For 137 clinical isolates tested by Spoligotyping, 104 are Mycobacterium tuberculosis with the sensitivity of 100.0%(104/104), 7 are Mycobacterium avium complex with the sensitivity of 77.8%(7/9), 15 are Mycobacterium intracellulare with the sensitivity of 93.8%(15/16), 1 are Mycobacterium fortuitum with the sensitivity of 0.0%(0/0), 10 are Mycobacterium chelonei with the sensibility of 100.0%(8/8) by DNA chip method, respectively. Kappa =0.95, U =30.6, P Conclusion The drug-sensitive test by DNA chips method matched the absolute concentration method very well. RFP-resistance was related to the mutations at the sites of 531, 516, 526, 511 for rpoB gene, and INH-resistance was related to the mutations at the site of 315 for KatG gene. No mutation was found in inhA gene in INH-resistant isolates. DNA chip method might be a rapid and effective method for the detection of Mtb drug-resistant isolates and mycobacterium species.

目的 应用基因芯片技术快速检测结核分枝杆菌的耐药基因型以及分枝杆菌菌种基因型，探讨基因芯片技术检测结核分枝杆菌耐药性和基因型的临床价值。 方法 应用菌种分型基因芯片技术和间隔区寡核苷酸分型（spacer oligonucleotide typing, Spoligotyping)技术对长沙市中心医院2011年1月至2012年12月的临床标本分离菌株137株进行菌种鉴定。对鉴定为结核分枝杆菌的菌株应用绝对浓度法进行利福平和异烟肼药物敏感性检测。并进一步对结核分枝杆菌临床分离株（利福平耐药、异烟肼耐药、敏感株）应用耐药基因芯片技术对 rpoB 、 katG 、 inhA 基因的野生型位点及各突变位点进行检测。即 rpoB 基因中1个或多个位点的碱基突变为利福平耐药， katG 、 inhA 基因任何基因中的1个或多个位点的碱基突变为异烟肼耐药。Spoligotyping法和基因芯片法菌种鉴定比较应用 Kappa 检验，绝对浓度法和基因芯片法药物敏感性比较采用χ 2 检验和 Kappa 检验，均以 P 结果 （1）与绝对浓度法比较，对利福平敏感和异烟肼敏感的菌株45株，耐药基因芯片法检测鉴定为敏感株、即野生型的 rpoB 基因42株，符合率为93.3%(42/45)（不符合的1株为511T-C突变，1株为531C-T突变, 1株为516A-T突变）； KatG 基因45株，符合率为100.0%(45/45)； inhA 基因43株，符合率为95.6%(43/45)（不符合的2株为inhA-15C-T突变）。（2）与绝对浓度法比较，对利福平轻度耐药的耐药菌株18株，耐药基因芯片检测鉴定为 rpoB 基因突变型的16株，符合率为88.9%(16/18)，且与531、516、526、511位点突变相关。对利福平高度耐药的耐药菌株19株，耐药基因芯片检测全部鉴定为 rpoB 基因突变型，符合率为100.0%(19/19)，且与531、516、526、511位点突变相关。(3)与绝对浓度法比较，对异烟肼轻度耐药的耐药菌株13株，耐药基因芯片检测鉴定为 KatG 基因突变型的12株，符合率为92.3%(12/13)，与315G-C和315G-A位点突变相关。 inhA 基因突变型0株。对异烟肼高度耐药的耐药菌株9株，耐药基因芯片检测鉴定为 KatG 基因全部为突变型，符合率为100.0%(9/9)，与315G-C和315G-A位点突变相关。 inhA 基因突变型0株。（4）与Spoligotyping法比较，137株临床分离株中104株分离株基因芯片法检测鉴定为结核分枝杆菌复合群，33株基因芯片法检测出7株鸟分枝杆菌、15株胞内分枝杆菌、1株偶然分枝杆菌、10株龟或脓肿分枝杆菌，与Spoligotyping法比较，结核分枝杆菌复合群一致性为100.0%（104/104），鸟分枝杆菌一致性为77.8%（7/9），胞内分枝杆菌一致性为93.8%（15/16），偶然分枝杆菌一致性为0.0%（0/0），龟或脓肿分枝杆菌一致性为100.0%（8/8）。经 Kappa 检验， Kappa =0.95, U =30.6, P 结论 耐药基因芯片检测法与绝对浓度法有高度的一致性，且利福平耐药与 rpoB 基因的531、516、526、511位点突变相关。异烟肼耐药与 KatG 基因的315位点突变相关，没有发现 inhA 相关的耐药位点突变。基因芯片方法可快速、准确地检测临床分离株的菌种基因型和耐药性。

Objective To analyze the influencing factors for sputum culture during sputum collection, storage and transport.  Methods The data of sputum culture from a municipal TB lab in Jiangsu province from April to October in 2010 were analyzed.  Results The process including sputum specimen collection, storage, transportation and culture in 475 smear-positive patients were analyzed, and found that the sputum specimens from different counties ( P =0.007), different specimen quantity ( P =0.008), different specimen storage time（ P =0.037）and different positive class （ P =0.001）all had statistic significance for culture quality.  Conclusions The procedure of specimen collection, storage, transportation and culture influence the result of culture. Their standard procedure required by the NTP should be strictly followed, and the training for health workers in county laboratory should be strengthened to improve sputum culture quality.

目的 通过分析某市级结核病实验室痰培养数据，总结县级运送阳性痰标本至市级进行痰培养工作模式对痰培养结果的影响因素。 方法 对江苏省某市级结核病实验室2010年4月—10月痰培养数据进行分析。 结果 该实验室初诊涂阳病人痰培养共计475例，对痰标本采集、储存、运输、培养等各个环节进行分析，得出：不同送样县区（ P =0.007），不同痰标本数量（ P =0.008），不同送样时间间隔（ P =0.037）,不同痰标本阳性级别（ P =0.001）的痰培养结果均有统计学意义。 结论 由县级运送痰标本至市级进行痰培养，应在采集标本时严格按照《中国结核病防治规划实施工作指南》（以下简称《指南》）要求采集即夜晨3份痰标本，提高送检痰标本质量，尽快送检，并加强对不同县区样品采集运送人员的培训。