—  BINFORD-DAMMIN SOCIETY OF INFECTIOUS DISEASE PATHOLOGISTS   —

Pediatric Mycobacterial Infections: Diagnostic Challenges and New Diagnostic Modalities


Gary W. Procop
Cleveland Clinic Foundation
Cleveland, OH


Tuberculosis and Non-tuberculous Infections of Children

Tuberculosis in Children

Congenital and Perinatal Tuberculosis
The diagnosis of congenital and perinatal tuberculosis is frequently delayed or missed. Approximately 50% of the infants born to mother with active tuberculosis will have the disease. A positive skin test on these neonates may not occur until 6 months of age. Tubercle bacilli from an infected mother may pass through the placenta to the fetus. Miliary tuberculosis may result from placental seeding. Infection may enter the fetus through the umbilical veins, which causes primary lesions in the liver with tuberculous lymph nodes in the porta hepatis. Lesions may also occur in the lungs, suggesting further passage of the bacilli through the ductus venosus. Tuberculous involvement of the lungs, and hilar and mediastinal lymph nodes, without associated hepatic lesions indicates aspiration of infected amniotic fluid or inhalation of tubercle bacilli or birth canal secretions

Congenital tuberculosis is characterized by a non-immune, non-reactive response, with abundant caseation and an unusually large number of bacilli. Polymorphonuclear leukocytes predominate, whereas lymphocytes, epithelioid histiocytes, and Langhans giant cells are rare.

Childhood Acquisition Lobar and segmental infections of the lung may occur when a young child contracts tuberculosis and the characteristic presentation is associated with the small diameter and compressibility of the airways. Historically, these lesions have been given the misleading term epituberculosis. These terms arose to contrast the apparent extent of the lung involvement, its relatively benign course, and the relative well-being of the child, in many cases. All of this suggested erroneously that the disease was not true tuberculosis. The name, however, is best supplanted by the term "segmental lesion".

The pathology consists of extrinsic bronchial obstruction from enlarged lymph nodes and intrinsic obstruction from edema and granulomas. Fistula formation with drainage of adjacent caseous lymph nodes into contiguous bronchi may also be present. The lobe or segments that are subtended by the obstruction may be the seat of obstructive hyperinflation, atelectasis, secondary non-tuberculous bacterial pneumonia, one or more isolated granulomas or aspiration of caseous material.

Atelectasis most commonly affects the anterior segments of the upper lobes and the right middle lobe, and in the pre-antibiotic era, full re-expansion of the affected segment was uncommon. Residual scarring was common in the absence of antibiotic therapy, as was residual bronchostenosis or bronchiectasis. Calcified residua of the primary complex, parenchymal, or lymph node lesions, may occasionally erode into a bronchus forming a broncholith, which may be expectorated or cause bronchial obstruction leading to distal infection and lung abscess.

Primary pulmonary disease in very young children is often asymptomatic, whereas older children and adolescence may develop an upper lobe infiltrate and progress to cavitary disease. If the primary infection is not contained, primary progressive tuberculosis ensues, with bronchopneumonia or lobar pneumonia, and cavitation. Primary progressive tuberculosis may meld into miliary tuberculosis, particularly in children less than three years of age. These children are also at risk for tuberculous meningitis. Primary disease that is severe or neglected may result in enlarged hilar and/or mediastinal lymph nodes that cause impaired venous return and superior vena cava syndrome, tracheal obstruction at the thoracic inlet, rupture into the mediastinum, erosion into a blood vessel or pericardium, compression of the left recurrent laryngeal or phrenic nerve, or compression of or erosion into the esophagus, leading to esophageal diverticulum and bronchoesophageal fistula.

Childhood Infections with Non-Tuberculous Mycobacteria (NTM)
Lymphadenitis - The most common presentation of a childhood infection with a NTM is lymphadenitis. This is usually unilateral and occurs in the cervical, submandibular or pre-auricular lymph node chains. It is most common in children less than five years old. In the past, the most common cause was M. scrofulaceum, but this mycobacteria has been replaced by members of the M. avium/M. intracellulare complex. Mycobacterium kansasii is also a rare cause of lymphadenitis in children.

Skin and Soft Tissue Infections – The so-called "fish tank granuloma" caused by M. marinum may occur in children with the appropriate aqueous exposure. Soft tissue infections with rapidly-growing mycobacteria may also occur.

Pulmonary Infections – Rare. When it occurs, it mimics tuberculosis and likely is caused by M . kansasii.

Disseminated Disease – Occurs in immunocompromised patients with AIDS or other severe immunodeficiencies. Members of the M. avium complex and M. intracellulare are the likely suspects.

Table 1. Cleveland Clinic Foundation Five Year Experience: 15 positive cultures

Mycobacteria (N) Ages/Sex Disease Associations
MAI complex 5 18M/M, 17M/F, 8Y/M, 8Y/M, 10Y/F, Parotid infection; cervical LA; common variable immunodeficiency, heart transplant/leiomyosarcoma, chronic heart/lung disease.
M. gordonae 3 13Y/M, 10Y, 21Y/M Non-pathogen/contaminants x 3
M. fortuitum 2 19Y/F, 16Y/M Ankle infectionChronic lung disease (Pul HTN)
M. tuberculosis 2 13Y/F, 22Y/F Cervical LA; cavitary pulmonary TB
M. marinum 1 15Y/F Fish tank granuloma
M. xenopi 1 24Y/M Possible skin infection associated with neutropenia
M. triplex 1 14Y/F Post-liver transplant

Diagnostic Tests: Current Testing Strategies and Assay Limitations.

Staining for Acid Fast Bacilli -
Staining followed by microscopic observation is applied to both direct smear examinations of patient specimens and organisms growing from the culture. If an organism is present in sufficient numbers, this is the most rapid procedure for the detection of mycobacteria in clinical samples. It has been estimated that at least 105 organisms/ml of sputum must be present to be detected by staining of a smear, and the sensitivity of the smear is related to the type of infection (i.e., advanced cavitary disease), relative centrifugal force used to concentrate the specimen, and other factors. Overall, the acid-fast smear alone used with clinical specimens is not an adequate diagnostic tool. Further, it does not provide information concerning the identification or viability of the organism, since all species of mycobacteria are acid-fast.

Two procedures are commonly used for acid-fast staining: carbol-fuchsin methods, including the Ziehl-Neelsen and Kinyoun procedures, and a fluorochrome method using auramine O or auramine-rhodamine dyes. The Ziehl-Neelsen and Kinyoun staining procedures differ in their staining principles, but both stain the mycobacterial cells red against a methylene blue counterstain. The stained smears must be viewed using a 100X oil-immersion objective. Auramine O-stained mycobacteria are bright yellow against a dark background and are easily visualized using a 25X objective. Modification of the auramine O technique includes the use of rhodamine, which gives a golden appearance to the cells. Auramine-rhodamine is the method of choice for clinical specimens, including tissue sections when trying to detect for mycobacteria.

Traditional Culture/HPLC/and Genetic Probes
Solid and broth media are used for the cultivation of mycobacteria. Mycobacteria grow more rapidly in the broth media, but colonial characteristics are not present which aid in identification. Generally, the recovery time for M. tuberculosis complex varies; on solid media, colonies can be observed in as short a time as 12 days or as long as 4 to 6 weeks with an average of about 3 to 4 weeks. The biochemical and biophysical properties of the mycobacteria are used for traditional identification. For example, the traditional identification of M. tuberculosis complexhas relied on acid-fastness, niacin production, nitrate reduction, and inactivation of catalase at 68º C.

The identification of mycobacteria in many laboratories has been supplemented with the used of high performance liquid chromatography (HPLC) and species-specific DNA probes, both of which are commercially available. The long-chained cell wall, fatty acids and mycolic acids extracted from mycobacteria are species-specific and yield different chromatographic peaks when separated by HPLC. The number of peaks, heights, and their position are used for identification of mycobacteria. Commercially-available genetic probes have been used since the beginning of the 1990's for the identification of mycobacteria. Presently probes (AccuProbe, Gen-Probe, Inc., San Diego, CA) are available for the M. tuberculosis complex, Mycobacterium kansasii, Mycobacterium avium/ M. intracellulare, and Mycobacterium gordonae. Although these are very useful, they only provide a definitive identification if the test is positive. If the test is negative, the user must try another probe and/or resort to traditional biochemical methods or another approach of identification.

New Diagnostic Testing of the Rapid Detection and Identification of Mycobacteria.

PCR and Real-Time PCR
There are two FDA-approved assays, and numerous laboratory validated ("home-brew") PCR and real-time PCR assays for the detection of M. tuberculosis [1, 2] .. These assays, although not a replacement for culture, may shorten the detection of M. tuberculosis in clinical specimens to hours from days-to-weeks. However, an important concern posed by the clinicians regarding these tests is that of false-negative results for M. tuberculosis. There would be significant infection control and possibly medicolegal implications if a person with active tuberculosis were taken out of respiratory isolation based on the erroneous M. tuberculosis PCR result.

Real-time PCR is a more rapid method of performing PCR wherein amplification and detection occur in the same reaction vessel. This reaction vessel is closed, which significantly decreases the chance of amplicon contamination of the laboratory. We have described an assay designed to avoid the possibility of a false-negative results on any smear-positive specimen. The assay is a broad-range PCR that detects all mycobacteria and differentiates M. tuberculosis from non-tuberculous mycobacteria based on post-amplification melt curve analysis (Figure 1) [3] . If a specimen that contains an acid fast bacillus is assayed with this PCR, the only diagnostic possibilities are: M. tuberculosis, non-tuberculous mycobacteria, or assay failure - which would be the result if PCR inhibition occurred. This assay was originally validated using 186 cultured isolates and 50 clinical specimens that were culture positive for M. tuberculosis. 48/50 specimens were detected by both this LightCycler assay and the FDA approved Amplicor M. tuberculosis PCR test to which it was compared. In addition to this assay, a variety of other excellent real-time PCR assays for M. tuberculosis have been described.



Figure 1. Post-amplification melt curve analysis may be used to differentiate M. tuberculosis from non-tuberculous mycobacteria.

Reverse Hybridization, DNA Sequencing and Microarrays

Reverse Hybridization
Reverse hybridization is similar in many was to the traditional Southern Blot, wherein the amplicon-probe hybridization reaction occurs on a nitrocellulose or similar substrate. In this technology, the multiple probes are immobilized on a nitrocellulose strip, and the amplicon is applied to the strip, which is the reverse of a Southern Blot. Lines or dots form at the site of amplicon-probe hybridization. When this pattern is compared with a key, one can interpret the results of this reaction. The advantage over traditional Southern blotting is that numerous probes are assayed simultaneously and radioisotopes are not used. This technology has been used successfully following broad-range PCR directed against the 16S to 23S spacer region to detect all the major pathogenic mycobacteria (Figure 2). Unfortunately, these assays may be cost-prohibitive and are not readily available in the United States.



Figure 2. The reverse hybridization line probe assay differentiates many clinically-important mycobacteria following PCR.

DNA Sequencing
DNA sequencing for the analysis of an amplified product is now a common method of post-amplification analysis. Although useful, this technology is more complicated than simple probe hybridization and often requires the user to have experience with sequence alignment and editing software, and genetic databases. Analysis of these variable regions interspersed between conserved regions that serve as broad-range primer hybridization sites is a powerful tool for microorganism identification. Traditional DNA sequencing (i.e. Sanger sequencing) was once a tool used solely in research laboratories, but has also become commonplace in many molecular pathology and molecular microbiology laboratories. These methods have been used successfully for the identification of bacteria, mycobacteria, Nocardia, and fungi [4, 5, 6, 7, 8, 9] . The genes that encode for the ribosomal subunits of these organisms are the most commonly used genetic targets the sequence-based identification. The short hypervariable regions that allow for a great degree of differentiation of these microorganisms have also been interrogated using a newer very user-friendly technology, pyrosequencing or sequencing by synthesis (Figure 3) [10]. We are in the process of implementing pyrosequencing as part of the routine identification of mycobacteria at the Cleveland Clinic Foundation. In addition to the 16S rDNA gene, many other genetic targets, such as rpoB and hsp have been used successfully for the sequence-based identification of mycobacteria. [6, 11, 12, 13, 14, 15, 16, 17] .



Figure 3. The short-read sequencing by synthesis analysis of the hypervariable region A of the 16S rDNA aids in the identification of mycobacteria.

Microarray Analysis
Microarrays, devices commonly referred to as gene chips, have been used extensively for research. These have also been used for the identification of mycobacteria and the detection of the genetic determinants of resistance. Troesch et al describe the use of a microarray that examined to genetic regions, the 16S rDNA and the rpoB gene [18]. They examined seventy mycobacteria representing 27 different species, and 15 rifampin-resistant isolates of M. tuberculosis with this microarray and were able to identify 26/27 species and all of the resistant mutants. Microarrays hold great promise, but are expensive and currently remain a research tool. Newer approaches to microarrays, such as bioelectronic arrays and liquid-based microarrays, with a limited probes for detection may bring microarray use into the clinicial laboratory very soon.

References

  1. Kraus, G., et al., Rapid and specific detection of the Mycobacterium tuberculosis complex using fluorogenic probes andreal-time PCR. Mol Cell Probes, 2001. 15(6): p. 375-83.

  2. Miller, N., et al., Rapid and specific detection of Mycobacterium tuberculosis from acid-fast bacillus smear-positive respiratory specimens and BacT/ALERT MP culture bottles by using fluorogenic probes and real-time PCR. J Clin Microbiol, 2002. 40(11): p. 4143-7.

  3. Shrestha, N.K., et al., Detection and differentiation of Mycobacterium tuberculosis and nontuberculous mycobacterial isolates by real-time PCR. J Clin Microbiol, 2003. 41(11): p. 5121-6.

  4. Cloud, J.L., et al., Evaluation of partial 16S ribosomal DNA sequencing for identification of nocardia species by using the MicroSeq 500 system with an expanded database. J Clin Microbiol, 2004. 42(2): p. 578-84.

  5. Mellmann, A., et al., Evaluation of RIDOM, MicroSeq, and Genbank services in the molecular identification of Nocardia species. Int J Med Microbiol, 2003. 293(5): p. 359-70.

  6. Khamis, A., et al., Usefulness of rpoB gene sequencing for identification of Afipia and Bosea species, including a strategy for choosing discriminative partial sequences. Appl Environ Microbiol, 2003. 69(11): p. 6740-9.

  7. Pauls, R.J., et al., A high proportion of novel mycobacteria species identified by 16S rDNA analysis among slowly growing AccuProbe-negative strains in a clinical setting. Am J Clin Pathol, 2003. 120(4): p. 560-6.

  8. Petrini, B., 16S rDNA sequencing in the species identification of non-tuberculous mycobacteria. Scand J Infect Dis, 2003. 35(8): p. 519-20.

  9. Hall, L., et al., Evaluation of the MicroSeq system for identification of mycobacteria by 16S ribosomal DNA sequencing and its integration into a routine clinical mycobacteriology laboratory. J Clin Microbiol, 2003. 41(4): p. 1447-53.

  10. Tuohy MJ, H.G., Procop GW. The Rapid Identification of Routine Clinical Mycobacteria by Pyrosequencing™. in 42nd ICAAC Meeting. 2002. San Diego, CA.

  11. Drancourt, M., et al., rpoB gene sequence-based identification of aerobic Gram-positive cocci of the genera Streptococcus, Enterococcus, Gemella, Abiotrophia, and Granulicatella. J Clin Microbiol, 2004. 42(2): p. 497-504.

  12. Drancourt, M. and D. Raoult, rpoB gene sequence-based identification of Staphylococcus species. J Clin Microbiol, 2002. 40(4): p. 1333-8.

  13. Fouad, A.F., et al., PCR-based identification of bacteria associated with endodontic infections. J Clin Microbiol, 2002. 40(9): p. 3223-31.

  14. Patel, J.B., et al., Sequence-based identification of Mycobacterium species using the MicroSeq 500 16S rDNA bacterial identification system. J Clin Microbiol, 2000. 38(1): p. 246-51.

  15. Ringuet, H., et al., hsp65 sequencing for identification of rapidly growing mycobacteria. J Clin Microbiol, 1999. 37(3): p. 852-7.

  16. Kwok, A.Y., et al., Species identification and phylogenetic relationships based on partial HSP60 gene sequences within the genus Staphylococcus. Int J Syst Bacteriol, 1999. 49 Pt 3: p. 1181-92.

  17. Pai, S., et al., Routine rapid Mycobacterium species assignment based on species-specific allelic variation in the 65-kilodalton heat shock protein gene (hsp65). Arch Pathol Lab Med, 1997. 121(8): p. 859-64.

  18. Troesch, A., et al., Mycobacterium species identification and rifampin resistance testing with high-density DNA probe arrays. J Clin Microbiol, 1999. 37(1): p. 49-55.