Contribution of the polymerase chain reaction to the diagnosis of tuberculous infections in children

A Bioinformatics Whole-Genome Sequencing Workflow for Clinical Mycobacterium tuberculosis Complex Isolate Analysis, Validated Using a Reference Collection Extensively Characterized with Conventional Methods and In Silico Approaches

The use of whole-genome sequencing (WGS) for routine typing of bacterial isolates has increased substantially in recent years. For Mycobacterium tuberculosis (MTB), in particular, WGS has the benefit of drastically reducing the time required to generate results compared to most conventional phenotypic methods. Consequently, a multitude of solutions for analyzing WGS MTB data have been developed, but their successful integration in clinical and national reference laboratories is hindered by the requirement for their validation, for which a consensus framework is still largely absent. We developed a bioinformatics workflow for (Illumina) WGS-based routine typing of MTB complex (MTBC) member isolates allowing complete characterization, including (sub)species confirmation and identification (16S, csb/RD, hsp65), single nucleotide polymorphism (SNP)-based antimicrobial resistance (AMR) prediction, and pathogen typing (spoligotyping, SNP barcoding, and core genome multilocus sequence typing). Workflow performance was validated on a per-assay basis using a collection of 238 in-house-sequenced MTBC isolates, extensively characterized with conventional molecular biology-based approaches supplemented with public data. For SNP-based AMR prediction, results from molecular genotyping methods were supplemented with in silico modified data sets, allowing us to greatly increase the set of evaluated mutations. The workflow demonstrated very high performance with performance metrics of >99% for all assays, except for spoligotyping, where sensitivity dropped to ∼90%. The validation framework for our WGS-based bioinformatics workflow can aid in the standardization of bioinformatics tools by the MTB community and other SNP-based applications regardless of the targeted pathogen(s). The bioinformatics workflow is available for academic and nonprofit use through the Galaxy instance of our institute at https://galaxy.sciensano.be.

Strong increase of true and false positive mycobacterial cultures sent to the National Reference Centre in Belgium, 2007 to 2016

Introduction

In 2007, a new federal legislation in Belgium prohibited non-biosafety level 3 laboratories to process culture tubes suspected of containing mycobacteria.

Aim

To present mycobacterial surveillance/diagnosis data from the Belgian National Reference Centre for mycobacteria (NRC) from 2007 to 2016.

Methods

This retrospective observational study investigated the numbers of analyses at the NRC and false positive cultures (interpreted as containing mycobacteria at referring clinical laboratories, but with no mycobacterial DNA detected by PCR in the NRC). We reviewed mycobacterial species identified and assessed trends over time of proportions of nontuberculous mycobacteria (NTM) vs Mycobacterium tuberculosis complex (MTBc), and false positive cultures vs NTM.

Results

From 2007 to 2016, analyses requests to the NRC doubled from 12.6 to 25.3 per 100,000 inhabitants. A small but significant increase occurred in NTM vs MTBc proportions, from 57.9% (587/1,014) to 60.3% (867/1,437) (p < 0.001). Although NTM infection notification is not mandatory in Belgium, we annually received up to 8.6 NTM per 100,000 inhabitants. M. avium predominated (ca 20% of NTM cultures), but M. intracellulare culture numbers rose significantly, from 13.0% (74/587) of NTM cultures in 2007 to 21.0% (178/867) in 2016 (RR: 1.05; 95% CI: 1.03–1.07). The number of false positive cultures also increased, reaching 43.3% (1,097/2,534) of all samples in 2016.

Conclusion

We recommend inclusion of NTM in sentinel programmes. The large increase of false positive cultures is hypothesised to result from processing issues prior to arrival at the NRC, highlighting the importance of sample decontamination/transport and equipment calibration in peripheral laboratories.

Role of PCR method using IS6110 primer in detecting Mycobacterium tuberculosis among the clinically diagnosed childhood tuberculosis patients at an urban hospital in Dhaka, Bangladesh

OBJECTIVE

Better methods are needed for the accurate detection of child tuberculosis (TB). This study compared different laboratory tests and evaluated IS6110 PCR for the detection of Mycobacterium tuberculosis (MTB) among clinically diagnosed child TB patients.

METHODS

A total of 102 paediatric patients (<15 years old) with clinically diagnosed TB were enrolled in this study. The patients were admitted to the icddr,b hospital in Dhaka between 2003 and 2005. Sputum/gastric lavage samples were collected for smear microscopy, culture (solid/Lowenstein-Jensen medium and liquid/MGIT), and IS6110 PCR testing. The sensitivity, specificity, and positive and negative predictive values (PPV, NPV) of smear microscopy and PCR were compared to the two culture methods.

RESULTS

Three patients were positive on smear microscopy (2.9%). MTB was detected by conventional culture in 15.7% (16/102), liquid culture in 14% (14/100), and IS6110 PCR in 61.8% (63/102). PCR detected an additional 45 patients who were undetected with the three other tests. Compared to conventional and liquid culture, respectively, smear microscopy showed sensitivity of 18.8% and 21.4%, specificity of 100% individually, PPV of 100% individually, and NPV of 86.9% and 88.7%, whereas PCR had sensitivity of 87.5% and 92.9%, specificity of 43% individually, PPV of 22.2% and 21%, and NPV of 94.9% and 97.4%.

CONCLUSIONS

PCR can be useful compared to smear microscopy and culture methods and is applicable as a rapid screening test for child TB. A larger scale study is required to determine its diagnostic efficacy in improving the detection of child TB in the presence and absence of severe malnutrition.

Matrix-Assisted Laser Desorption Ionization-Time of Flight Mass Spectrometry for Combined Species Identification and Drug Sensitivity Testing in Mycobacteria

Species identification and drug susceptibility testing (DST) of mycobacteria are important yet complex processes traditionally reserved for reference laboratories. Recent technical improvements in matrix-assisted laser desorption ionization-time of flight mass spectrometry (MALDI-TOF MS) has started to facilitate routine mycobacterial identifications in clinical laboratories. In this paper, we investigate the possibility of performing phenotypic MALDI-based DST in mycobacteriology using the recently described MALDI Biotyper antibiotic susceptibility test rapid assay (MBT-ASTRA). We randomly selected 72 clinical Mycobacterium tuberculosis and nontuberculous mycobacterial (NTM) strains, subjected them to MBT-ASTRA methodology, and compared its results to current gold-standard methods. Drug susceptibility was tested for rifampin, isoniazid, linezolid, and ethambutol (M. tuberculosis, n = 39), and clarithromycin and rifabutin (NTM, n = 33). Combined species identification was performed using the Biotyper Mycobacteria Library 4.0. Mycobacterium-specific MBT-ASTRA parameters were derived (calculation window, m/z 5,000 to 13,000, area under the curve [AUC] of >0.015, relative growth [RG] of <0.5; see the text for details). Using these settings, MBT-ASTRA analyses returned 175/177 M. tuberculosis and 65/66 NTM drug resistance profiles which corresponded to standard testing results. Turnaround times were not significantly different in M. tuberculosis testing, but the MBT-ASTRA method delivered on average a week faster than routine DST in NTM. Databases searches returned 90.4% correct species-level identifications, which increased to 98.6% when score thresholds were lowered to 1.65. In conclusion, the MBT-ASTRA technology holds promise to facilitate and fasten mycobacterial DST and to combine it directly with high-confidence species-level identifications. Given the ease of interpretation, its application in NTM typing might be the first in finding its way to current diagnostic workflows. However, further validations and automation are required before routine implementation can be envisioned.

Tuberculosis diagnosis using immunodominant, secreted antigens of Mycobacterium tuberculosis

Tuberculosis (TB) remains a major public health concern in most low-income countries. Hence, rapid and sensitive TB diagnostics play an important role in detecting and preventing the disease. In addition to established diagnostic methods, several new approaches have been reported. Some techniques are simple but time-consuming, while others require complex instrumentation. One prominent and readily available approach is to detect proteins that Mycobacterium tuberculosis secretes, such as Mpt64, the 6-kDa early secreted antigenic target (Esat6), the 10-kDa culture filtrate protein (Cfp10), and the antigen 85 (Ag85) complex. Although their functions are not fully understood, a growing body of molecular evidence implicates them in M. tuberculosis virulence. Currently these biomarkers are either being used or investigated for use in skin patch tests, biosensor analyses, and immunochromatographic, immunohistochemical, polymerase chain reaction-based, and enzyme-linked immunosorbent assays. This review provides a comprehensive discussion of the roles these immunodominant antigens play in M. tuberculosis pathogenesis and compares diagnostic methods based on the detection of these proteins with more established tests for TB.

Tuberculosis in childhood

There has been a recent global resurgence of tuberculosis (TB) fuelled by HIV infection and migration. Childhood TB represents a sentinel event in the community, suggesting recent transmission from an infectious adult. The diagnosis of TB in children is based on chest X-ray, tuberculin skin testing and mycobacterial staining/culture, although the diagnostic yield from these investigations is often lower than in adults. Newer diagnostic tests are being developed and may improve the diagnostic yield in childhood TB. Treatment of TB in children is similar to adults in that short-course multidrug treatment has been adopted as standard therapy in many national TB programmes. Compliance is a major determinant of the success of drug treatment and directly observed therapy has been adopted as a key component of TB treatment programmes. Although uncommon in children, multidrug-resistant TB is also increasing and treatment often involves longer courses of therapy with second-line drugs.

[Specific features of tuberculosis in childhood]

Primary infection with Mycobacterium tuberculosis usually occurs during childhood. The source of infection is most often an adult. The risk of infection in exposed children is modulated by various factors related to the infectiousness of the index case, exposure conditions, and the child himself. This review aims to describe the specific diagnostic and therapeutic features of latent TB infection and TB disease in childhood.

New specimens and laboratory diagnostics for childhood pulmonary TB: progress and prospects

Childhood pulmonary TB (PTB) is under diagnosed, in part due to difficulties in obtaining microbiological confirmation. However, given the poor specificity of clinical diagnosis, microbiological confirmation and drug susceptibility testing is important in guiding appropriate therapy especially in the context of drug resistant TB. Confirmation is often possible, even in infants and young children, if adequate specimens are collected. Culture yield varies with the severity of illness, specimen type and culture method. Induced sputum is recognised as a safe procedure with a high diagnostic yield. Advances include optimised protocols for smear microscopy and modified culture techniques, such as the Microscopic Observation Drug Susceptibility Assay. Detection of Mycobacterium tuberculosis nucleic acid in respiratory specimens has high specificity but relatively poor sensitivity, particularly for smear negative disease. The recent development of an integrated specimen processing and real-time PCR testing platform for M. tuberculosis and rifampicin resistance is an important advance that requires evaluation in childhood TB.

Diagnosis of pulmonary tuberculosis in children: new advances

The global burden of childhood pulmonary TB has been underappreciated, in part due to difficulties in obtaining microbiological confirmation of disease. Most HIV-uninfected children can be diagnosed using a combination of clinical and epidemiological features, tuberculin skin testing and chest radiography, as represented in different scoring systems. However, accurate microbiologic diagnosis has become increasingly important for timely use of effective treatment. Mycobacterial culture confirms the diagnosis of TB and provides drug susceptibility data but is not available in most areas with a high TB prevalence. Moreover, culture has poor sensitivity in children who usually have paucibacillary disease. The HIV epidemic has made definitive diagnosis even more challenging due to nonspecific clinical and radiological signs. In high HIV-prevalence areas, scoring systems have been especially variable, lacking sensitivity and specificity. Newer methods for diagnosis are aimed either at detecting the organism or a specific host immune response. Methods for organism detection have focused on collection of better samples, improved culture techniques, molecular methods or antigen detection. Recent advances include the use of sputum induction for obtaining a more reliable specimen, faster and more sensitive culture methods, and rapid detection of the organism and drug resistance based on nucleic acid amplification. Improved methods for detecting a specific host response have largely focused on the use of IFN-g release assays. Even with newer methods, accurately diagnosing childhood TB may be challenging. Greater efforts to obtain a microbiologic diagnosis should be made in children, even in primary care settings. Further research to develop a more accurate, cost-effective and simple diagnostic test for childhood TB is urgently needed.

Microbiological diagnostic procedures in respiratory infections: mycobacterial infections

This article will review traditional and newer microbiological techniques for the diagnosis of mycobacterial respiratory infections. It will concentrate on the diagnosis of infections due to Mycobacterium tuberculosis, the main mycobacterium causing respiratory infections of clinical and public health importance. The diagnosis of respiratory disease associated with non-tuberculous mycobacteria (NTM), particularly in children with underlying airway pathology such as cystic fibrosis (CF) or bronchiectasis, will be briefly discussed. With respect to the diagnosis of tuberculosis (TB), the review will concentrate on the diagnosis of patients with symptoms and/or signs of clinical disease, rather than the detection of exposure or asymptomatic infection. It will not specifically address the assessment of pre-test probability based on clinical or epidemiological factors, the use of radiological investigations or the investigation of extrathoracic lymph node disease or chest wall disease. The role of newer diagnostic modalities including nucleic acid detection (NAD) and gamma-interferon assays in paediatric practice will be reviewed, and suggestions made as to how they may fit into contemporary diagnostic algorithms.

Utility of fast mycobacterial interspersed repetitive unit-variable number tandem repeat genotyping in clinical mycobacteriological analysis

BACKGROUND

Analysis of variable numbers of tandem repeats (VNTR) of genetic elements called mycobacterial interspersed repetitive units (MIRUs) is a recently described, polymerase chain reaction (PCR)-based method used to genotype Mycobacterium tuberculosis. It is much faster, requires a smaller amount of DNA, and has approximately the same discriminatory power as the standard IS6110 restriction fragment-length polymorphism (RFLP) method. We report the adaptation and optimization of MIRU-VNTR genotyping on a capillary electrophoresis system. We describe its application to 3 typical clinical situations encountered in our laboratory (Institut Pasteur de Bruxelles, Laboratoire Tuberculose et Mycobacteries; Brussels, Belgium).

METHODS

MIRU-VNTR genotyping was performed on heat-inactivated M. tuberculosis cultures obtained from clinical specimens on Lowenstein solid medium or in mycobacteria growth indicator liquid tubes (Becton Dickinson). After amplification of 12 genomic loci using 4 different multiplex PCRs, DNA fragments were separated by capillary electrophoresis using the ABI Prism 3100-Avant Genetic Analyzer (Applied Biosystems). Sizing of the PCR fragments and assignment of the various MIRU-VNTR alleles were done using the GeneScan and customized Genotyper software packages (PE Applied Biosystem).

RESULTS

Clustering on the basis of IS6110 fingerprinting of isolates from 3 different patients attending the same hospital was confirmed by MIRU-VNTR typing. This concordance between 2 independent, highly discriminatory techniques was decisive in triggering an epidemiological inquiry that led to identification of a bronchoscopy-related tuberculosis nosocomial infection. A mixed tuberculosis infection in a patient whose infection was initially suspected as a result of the IS6110 RFLP method was clearly identified by MIRU-VNTR typing. Finally, automated MIRU-VNTR analysis permitted the identification of laboratory contamination in 6 liquid cultures of M. tuberculosis within several hours.

CONCLUSION

These examples illustrate the utility of this genotyping technique for quick and accurate resolution of problems commonly encountered in clinical mycobacteriology.

Newer diagnostic modalities for tuberculosis

The gold standard for diagnosis of tuberculosis is demonstration of mycobacteria from various body fluids. This is often not possible in children due to pauci-bacillary nature of illness. Significant improvement in understanding of molecular biology of Mycobacterium tuberculosis has led to development of newer diagnostic techniques of tuberculosis. Polymerase chain reaction (PCR) is an emerging diagnostic tool for diagnosis of TB in children. However, its role in day-to-day clinical practice needs to be defined. A negative PCR never eliminates possibility of tuberculosis, and a positive result is not always confirmatory. The PCR may be useful in evaluating children with significant pulmonary disease when diagnosis is not readily established by other means, and in evaluating immunocompromised children (HIV infection) with pulmonary disease. In the absence of good diagnostic methods for tuberculosis, a lot of interest has been generated in serodiagnosis. ELISA has been used to detect antibodies to various purified or complex antigens of M. tuberculosis in children. Despite a large number of studies published over the past several years, serology has found little place in the routine diagnosis of tuberculosis in children, even though it is rapid and does not require specimen from the site of disease. Sensitivity and specificity depend on the antigen used, gold standard for the diagnosis of tuberculosis and the type of tubercular infection. Though most of these tests have high specificity, their sensitivity is poor. In addition, these tests may be influenced by factors such as age, prior BCG vaccination and exposure to environmental mycobacteria. The serological tests, theoretically, may not be able to differentiate between infection and disease. At present, serodiagnosis does not appear to have any role in diagnosis of childhood pulmonary tuberculosis. A new test (QuantiFERON-TB or QFT) that measures the release of interferon-gamma in whole blood in response to stimulation by purified protein derivative is comparable with the tuberculin skin testing to detect latent tubercular infection, and is less affected by BCG vaccination. It can also discriminate responses due to nontuberculous mycobacteria, and avoids variability and subjectivity associated with placing and reading the tuberculin skin test. Polymerase chain reaction based test for identification of katG and rpoB mutation which are associated with isoniazid and rifampicin resistance may help in early identification of drug resistance in mycobacterium.

Diagnosis and treatment of tuberculosis in children

There has been a recent global resurgence of tuberculosis in both resource-limited and some resource-rich countries. Several factors have contributed to this resurgence, including HIV infection, overcrowding, and immigration. Childhood tuberculosis represents a sentinel event in the community suggesting recent transmission from an infectious adult. The diagnosis of tuberculosis in children is traditionally based on chest radiography, tuberculin skin testing, and mycobacterial staining/culture although these investigations may not always be positive in children with tuberculosis. Newer diagnostic methods, such as PCR and immune-based methods, are increasingly being used although they are not widely available and have a limited role in routine clinical practice. Diagnostic approaches have been developed for use in resource-limited settings; however, these diagnostic methods have not been standardised and few have been validated. Short-course, multidrug treatment has been adopted as standard therapy for adults and children with tuberculosis, with or without directly observed therapy. Compliance is a major determinant of the success of drug treatment. Although uncommon in children, multidrug-resistant tuberculosis is also increasing and treatment will often involve longer courses of therapy with second-line antituberculosis drugs. Treatment of latent infection and chemoprophylaxis of young household contacts is also recommended for tuberculosis prevention, although this may not always be carried out, particularly in high incidence areas.

Tuberculosis in children-is PCR the diagnostic solution?

Polymerase chain reaction (PCR) has been recently incorporated as a diagnostic tool for the diagnosis of tuberculosis. The benefit of rapid results and greater sensitivity compared with traditional microbiological methods makes PCR a suitable technique in childhood tuberculosis, especially when diagnosis is difficult or when urgent diagnosis is needed. However, the possibility of false-positive results must be considered, especially if the clinical and epidemiologic context of the child make the diagnosis of tuberculosis improbable. The commercial 'Amplicor PCR test' lacks good sensitivity and specificity and it would be necessary to develop other commercial easy-to-use PCR kits that provides better yield.

Comparison of amplicor, in-house polymerase chain reaction, and conventional culture for the diagnosis of tuberculosis in children

A total of 251 clinical specimens (235 gastric aspirates and 16 bronchoalveolar lavages) from 88 children were prospectively tested in a blinded manner for the presence of Mycobacterium tuberculosis complex, by use of the Amplicor M. tuberculosis test and by means of in-house polymerase chain reaction (PCR). The results were compared with those obtained by conventional culture and by direct microscopy. All of the children underwent extended follow-up to verify or exclude the clinical diagnosis of tuberculosis. The results of the different tests, when compared to the final clinical diagnosis, were a sensitivity of 60% and a specificity of 96.8% for in-house PCR, 44% and 93.7% respectively for the Amplicor test, 44% and 100% for mycobacterial culture and 12% and 100% for microscopy. Amplicor tests presented false-positive findings in children without tuberculous infection. We conclude that both in-house PCR and the Amplicor test are rapid methods that can be helpful for difficult or urgent diagnosis of tuberculosis in children. However, efforts should be aimed toward improvement of the sensitivity and specificity of an easy-to-use PCR kit.

Mycobacterium intracellulare as a cause of a recurrent granulomatous tenosynovitis of the hand

We report a case of recurrent granulomatous tenosynovitis with M. intracellulare in a 55-year-old HIV negative diabetic woman. Identification of the causative agent further than belonging to the M. avium-intracellulare complex is provided by specific PCR-amplification of genomic DNA and sequencing of an hypervariable region within its 16S RNA gene. Sixteen months antibiotic regimen of rifabutin and clarithromycin led to a complete resolution of the tenosynovitis.