The identification of latent tuberculosis infection (LTBI) in any individual or population has proven to carry significant importance not only for that person’s health, but also for the control and eventual elimination of tuberculosis (TB) in the United States. Traditionally, the tuberculin skin test (TST) has served as the standard of care for the identification of prior exposure to Mycobacterium tuberculosis (MTB).
However, the specificity of a positive test is less than optimal. It is either due to previous vaccination intended to prevent TB or infection with nontuberculous mycobacterium (NTM). Newer tests classified as interferon-gamma release assays (IGRA) possess potential advantages over the TST when used for identifying those with MTB infection. We recently diagnosed a case of pleuropulmonary infection involving an unusual NTM, Mycobacterium interjectum (M. interjectum), in an immunocompromised man diagnosed 1 year after he had been treated for LTBI based on a reactive TST. A propos this experience, we discuss the beneficial role of IGRAs and review the literature on infection with M. interjectum.
Historically, the tuberculin skin test (TST) has been relied upon for diagnosing persons who have been sensitized by Mycobacterium tuberculosis (MTB), a condition referred to as latent tuberculosis infection (LTBI).1 In a TST survey conducted in 2000, an estimated 4.2% of the civilian, non-institutionalized U.S. population aged >1 year had LTBI. Although this represented a 60% decline from 1972,2 the decrease in prevalence was not uniform across all of the segments of the population. Furthermore, approximately 9.4% of the 153,555 persons diagnosed with active tuberculosis (TB) during the 10-year period between 1998 and 2007 died either before treatment began or during therapy prior to completing the anti-tuberculous regimen. Because the rates of MTB infection and active TB vary considerably, targeted testing and the selection of those persons likely to benefit from treatment for latent infection (ie, persons who are at increased risk for a poor clinical outcome if active infection ensues), have been assigned a high priority.
Utilizing an intradermal injection of a polyvalent antigenic mixture labeled purified protein derivative (PPD), TST is designed to assess in vivo delayed-type hypersensitivity (DTH) (Type IV), thereby helping to identify those individuals who have been infected with MTB. The reliability of the TST may be influenced by its technical application and the need for a second encounter 48–72 hours after test administration in order to record any observed reaction. Currently the standard method for detecting LTBI, TST has limitations in regards to sensitivity and specificity. Conditions such as advanced age, immunocompromised states, prior vaccination with Bacillus Calmette-Guerin (BCG), and previous or current infection with nontuberculous Mycobacterium species (NTM), may hinder the interpretation of TST results.3
Accordingly, improved diagnostic tests to aid in the management of MTB disease have been approved by the U.S. Food and Drug Administration (FDA). Collectively known as interferon-gamma (IFN-γ) release assays (IGRAs), these novel laboratory tools have the potential advantage of heightened sensitivity and specificity, less interpreter bias, and test result availability in less than 24 hours. Additionally, the more recent of the IGRAs, QuantiFERON-TB Gold and Quanti FERON-TB GOLD In-Tube (QFT-G and QFT-GIT, Cellestis Limited, Carnegie, Victoria, Australia) and T-SPOT.TB (T-SPOT, Oxford Immunotec Limited, Abingdon, United Kingdom), are indicated for use in the diagnosis of latent as well as active infection with MTB.4,5
Recently, we encountered a man diagnosed with a pleural space infection involving Mycobacterium interjectum (M. interjectum) who previously had been treated with 9 months of isoniazid (INH) for LTBI following the discovery of a positive TST (12 mm of induration), performed in anticipation of treatment of an underlying rheumatologic condition with a tumor necrosis factor alpha (TNF-α) inhibitor. At the time of his positive TST, he was asymptomatic and was found to have a clear chest radiograph (CXR). A repeat CXR obtained 1 year later revealed new asymptomatic unilateral pleural thickening with an associated pleural effusion, subsequently confirmed by computerized tomographic (CT) scan imaging. His history of previous asbestos exposure prompted a thoracentesis in order to rule out the possibility of an occult mesothelioma. Upon cytologic examination, the exudative fluid was void of any cells suspicious for malignancy. However, microbiologic laboratory stains revealed the presence of acid-fast bacilli (AFB) subsequently identified as M. interjectum. Based on in vitro susceptibility test results, clarithromycin and rifabutin were administered for 6 months, during which time treatment with infliximab was discontinued. At clinical follow-up, he remained asymptomatic. However, no test of cure microbiologic data was made available. A propos this case, we discuss the potential role that an IGRA may have played in his earlier management, assuming the possibility of a false-positive TST elicited by infection with this uncommon NTM. In addition, we review the medical literature pertaining to infection with M. interjectum.
Presently, TB is the second most common cause of death from an infectious disease worldwide, accounting for approximately 2 million TB lives lost each year. The estimated one-third of the world’s population who harbor this mycobacterium and remain asymptomatic are collectively referred to as having LTBI. Identification of individuals with LTBI who are at high risk for developing active TB infection is of importance for the specific individual as well as for the control and potential elimination of TB disease globally. Included among the group considered to be at increased risk for active disease are those who have had recent infection with MTB as well as patients who have certain clinical conditions known to be associated with a heightened risk for progression of LTBI to active TB. Included among these conditions are chronic inflammatory diseases such as rheumatoid arthritis (RA), and the effects of immunosuppressive therapy directed at controlling this potentially crippling disease.
Commonly used medications such as corticosteroids, methotrexate (MTX), and other nonbiologic disease-modifying antirheumatic drugs (DMARDs) have been considered to contribute to immunosuppression.6 The TB targeted screening and treatment statement issued in 2000 by the U.S. Centers for Disease Control and Prevention (CDC) states that prednisone treatment at doses of 15 mg/day for more than 1 month represents a risk for TB.7 Low-dose MTX is the most widely used first line DMARD in the treatment of RA and is also used as the backbone in treatment with newer biological anti-TNF agents. Serving as a folate antagonist by interfering with purine/pyrimidine synthesis, MTX has been shown to decrease antigen stimulated T cell proliferation, neutrophil chemotaxis, and superoxide generation, as well as to decrease the release of TNF-α and INF-γ from T cells in patients with RA. Methotrexate is believed to not only increase one’s risk of infection, including MTB, but also intensify disease severity if and when infection occurs. However, contrary to published data involving corticosteroid medications, no large observational or prospective studies demonstrating a relationship of MTX to infection have firmly confirmed this position.8,9
Tumor necrosis factor-α is expressed by activated macrophages and T and B cell lymphocytes and is integral to granuloma formation and maintenance. It activates macrophages to ingest and kill mycobacteria and other pathogens.10 Although TNF-α has been demonstrated to play an essential role in host defense against infection in animal models, an independent association increasing patients’ risk for serious infections attributed to the use of TNF-α inhibitors (ie, infliximab, adalimumab, certolizumab pegol, afelimomab, golimumab, etanercept, and pegsunercept) has been difficult to define.6 Nevertheless, the reactivation of LTBI has become recognized as a well-documented risk in patients receiving TNF-antagonist therapy.11 Interestingly, data from RA patients in the Swedish Biologics Register along with other published investigations have suggested the risk for infection varies with the duration of exposure to the anti-TNF-α therapy.12,13 Infections of the respiratory tract and pneumonia were the most common, with risk being the greatest in the first 6 months after initiating anti-TNF therapy, particularly for infliximab users. Subsequently, the CDC has published guidelines to prevent TB in patients initiating TNF-α inhibitor therapy, including screening these subjects for TB risk factors and for LTBI prior to beginning anti-TNF medications.11
Beginning in 1962 with the observation that INH was effective in preventing TB among household contacts of persons with TB disease,14 investigations of contacts and treatment of those with LTBI became a strategy for the control and elimination of TB in the Unites States. Up until recently, TST was the only proven method for identifying infection with MTB in persons who did not have active TB disease. The immunologic basis for TST rests on the knowledge that infection with MTB produces a delayed-type (cellular) hypersensitivity (DTH) reaction to certain antigenic components (tuberculins) that are contained in an extract of culture filtrate of the organism. Initially made by Siebert in 1934 and remaining the preferred agent used for TST, PPD is isolated from this culture filtrate by protein precipitation. Using the standard method referred to as the Mantoux test, 0.1 mL of solution containing 5-tuberculin units of PPD is injected intracutaneously into the volar surface of the arm. Any induration appearing at the injection site after 48–72 hours is considered a reaction and is measured in millimeters (mm) and recorded. The proper interpretation of skin test reactions depends on TST sensitivity and specificity as well as the positive predictive value of a reactive test. Tuberculin skin test sensitivity is estimated to be nearly 100% in those with LTBI and possessing normal immune responsiveness. False-positive tests can occur in persons infected with NTM or in those previously vaccinated with BCG, thereby resulting in a lower specificity and positive predictive value in those individuals who have a low probability of LTBI. Consequently, regardless of how high of specificity the TST may have, testing low prevalence populations will statistically result in most reactors being false positives. In an attempt to overcome this problem, by progressively increasing the reaction size (eg, >5 mm, >10 mm, and >15 mm of induration) used to separate positive from negative tests, TST specificity is bolstered. Alternatively, false-negative TST results are known to occur in at least 20% of otherwise healthy persons with active TB, though most revert to positive shortly after effective treatment is initiated.15 Furthermore, TST may be falsely negative in those with protein malnutrition, sarcoidosis, intercurrent viral infections (eg, HIV with diminished CD4+ T cell lymphocyte counts), and reticuloendothelial diseases. Systemic corticosteroid therapy and certain biological modifying agents may also attribute to false-negative TST. These limitations, along with varying intra-observer reliability in reading skin test reactivity, make obvious the need for dependable and reliable alternative methods to help diagnose LTBI.
In 2001 the FDA approved the original IFGA, Quanti FERON-TB (QFT), which was designed to be used as an aid in diagnosing MTB infection and was indicated for LTBI only.16 However, QFT specificity was less than TST despite the use of Mycobacterium avium (M. avium) antigen as a control for nontuberculous mycobacterial sensitization and saline as a negative control.17 No longer commercially available, it was replaced in 2005 with the QFT-G test, the first in vitro available tool approved for the diagnosis of latent as well as active MTB infection. The QFT-G was recommended to be used in all circumstances where TST was indicated. However, its use was hampered by the need for a fresh blood specimen containing viable WBCs, thus being limited to laboratories possessing trained personnel capable of testing blood within a few hours after collection. Subsequently, in 2007 the FDA approved QFT-GIT, a modification of the previous QFT-G assay with the added advantage of providing extended transport time with virtually no labor, since the special tubes used to collect the blood contained antigens, and control materials allow for more direct testing of fresh blood. Using fresh heparinized whole blood, this enzyme-linked immunosorbent assay (ELISA) detects the release of IFN-γ from sensitized WBCs after incubation with test antigens. The amount of IFN-γ released in the plasma is determined after taking into account the background level of this cytokine. Results are reported as positive, negative, or indeterminate. Indeterminate tests do not provide useful information regarding the likelihood of MTB infection but should be repeated in individuals who have an increased risk of infection. On the contrary, no further tests are necessary following an indeterminate result in low-risk persons. Like QFT-G, the antigens used in the QFT-GIT test are a mixture of synthetic peptides simulating 2 proteins present in MTB: early secretory antigenic target-6 (ESAT-6) and culture filtrate protein-10 (CFP-10), both of which are secreted by all MTB and pathogenic Mycobacterium bovis strains. In addition, the QFT-GIT assay also includes a third MTB protein named TB7.7(p4), which was not included in the QFT-G assay. Since these proteins are absent from all BCG vaccine strains and commonly encountered NTM, except for M. kansasii, M. szulgai, and M. marinum, QFT-GIT is more specific for MTB than the TST, which uses PPD as the antigen.18 Other advantages of this IGRA include result availability <24 hours after testing without the need for a second visit; elimination of biases and errors of TST placement and reading; and no triggering of an anamnestic response (ie, booster effect), since it does not expose persons to antigen. Errors in collecting and transporting blood specimens as well as in running or interpreting the assay may decrease the accuracy of the QFT-GIT test. The most recent of the IGRAs, T-SPOT was FDA approved in 2008. Employing 2 amino acid sequence peptides in their entirety (ESAT-6 and CFP-10, along with control materials), this enzyme-linked immunospot assay (ELISpot) detects increases in the number of incubated peripheral blood mononuclear cells (PBMCs) that secrete IFN-γ after stimulation with antigen as compared to the media control.
As with other indirect tests for MTB infection (ie, TST), the accuracy of QFT-GIT and T-SPOT are hindered by the lack of tests to confirm a diagnosis of latent or active MTB infection in lieu of a positive culture. Published studies vary in their estimates of TST sensitivity compared with those of IGRAs. However, in general pooled estimates, although not consistent across tests and populations, suggest that TST is probably as sensitive as QuantiFERON but less sensitive than T-SPOT for the detection of LTBI.19 However, the performance of QFT-GIT and T-SPOT, based on sensitivity, have not been as well defined in patients suffering from impaired immune function, including HIV/AIDS; immunosuppressive medications (eg, high-dose corticosteroids, TNF-α, post-organ transplant drugs); specific malignancies and hematologic disorders; chronic renal failure; silicosis; and diabetes mellitus. Data are also lacking for pregnant women and individuals less than 5 years of age. On the other hand, greater specificity is an undeniable advantage of QFT-GIT and T-SPOT, but as with other diagnostic tests, the predictive value of results with either of these IGRAs strongly depends on the prevalence of MTB infection in the population being tested. Although each IGRA has its own characteristics, pooled specificity data obtained from various studies containing BCG-vaccinated and non-BCG-vaccinated samples demonstrated that regardless of the assay employed, specificity results were uniformly high.19 Data using IGRAs to confirm or refute infection with NTM are not available. A Dutch study has observed that among control subjects with a TST result of >10 mm, 44.4% were positive by QFT-G compared with 11.5% of participants returning from foreign missions.20 The discordant results were believed to be due to a large amount of false-positive TST reactions from exposure to NTM during missions to areas endemic for mycobacterium species other than TB. Although approved by the FDA for diagnosing LTBI and active TB disease, neither QFT-GIT nor T-SPOT can differentiate between the 2. Therefore, a medical evaluation would be necessary before declaring that any individual with a positive test has only LTBI. Historically, the lifetime risk for developing active TB has been estimated to be 5%–10% in a person with a positive TST. To date, there exists limited long-term data on the ability of any 1 IGRA to predict risk for subsequent active disease with MTB.
QuantiFERON-TB Gold In-Tube and T-SPOT may be used in place of the TST in all circumstances in which the latter is currently indicated. Situations in which an IGRA may be preferred but a TST is acceptable include testing of persons whom may not be relied on to return for TST readings as well as in those who have been previously vaccinated with BCG. Conversely, preference is given to TST over IGRAs for testing children aged <5 years. In general, other than during comparison studies of TST and IGRAs, there is no need to use both tests together except in a few special circumstances.5 These may include when either an initial test is negative in high-risk individuals or a strong clinical suspicion for active TB exists and confirmation of MTB infection is desired; when either initial test is positive and more evidence of infection is sought to promote compliance; or in healthy persons who are unlikely to have infection or progress if they are infected.
The term “nontuberculous mycobacteria” refers to a group of environmental mycobacterial species differing from those in the MTB complex and the Mycobacterium leprae (M. leprae group). Also referred to as “mycobacteria other than tuberculosis (MOTT)” and “atypical mycobacteria,” these environmentally ubiquitous organisms were once dismissed as contaminants when isolated from clinical specimens. The number and taxonomy of the NTM remained relatively constant for many years after they first came to be recognized as pathogens in the mid-1950s. However, in the past decade or so, the number of new species added to the list of NTM has increased dramatically.21 Increased recognition and knowledge of the NTM by the laboratory and clinician serve to strengthen the once uncertain significance of these newly described or emerging species. Previous dependence on traditional phenotypic methods for species identification, including cultural and biochemical analyses, has now been replaced by modern genetic techniques for the characterization of mycobacteria. In particular, examination of the 16S rRNA gene sequence allows precise determination of the isolated mycobacterial strain allowing for comparison with known species and identification of new taxa within the NTM.22 Traditionally, immunocompromised patients have been associated with a greater number of infections with mycobacterium other than MTB; however, many of the more recently identified uncommon NTM diseases are now also being diagnosed at a higher rate in normal hosts.
Mycobacterium interjectum is a scotochromogenic (forms yellow-orange pigment independent of light exposure) mycobacteria so named for its place between rapidly and slowly growing mycobacteria. Since its original description in 1993 in a young child with chronic lymphadenitis23 about 14 cases of infection involving this rare mycobacteria have been reported in the medical literature, almost all of which have involved lymph nodes (all pediatric cases) or the respiratory tract (all adult cases).24–26 In only 2 of the pulmonary cases were the M. interjectum isolates believed to be clinically significant. Susceptibility data and treatment recommendations are limited. Generally resistant to first line anti-mycobacterial agents, M. interjectum strains have been found to be susceptible to clarithromycin, rifabutin, and amikacin. With an unknown duration of therapy, a cure was achieved in most lymphadenitis cases only after total resection of the involved lymph node.
Our patient’s lack of access to the IGRA tests makes it impossible for us to declare with certainty that his reaction to the TST, performed prior to beginning treatment with a TNF-α inhibitor, was a false positive. We believe the development of the pleural effusion on the CXR, although an asymptomatic finding, was directly related to an intrathoracic infection with M. interjectum. Supporting this assertion are the absence of any effusion on the prior CXR at the time of his positive TST, exudative nature of the pleural fluid, identification of the mycobacterium by AFB staining and culture, and recognition of this NTM causing lung disease, albeit rare, in previous reports.
Presumably, medication-induced lowered immune function allowed for this likely latently harbored, normally indolent mycobacterium species to become pathogenic, thus leading to active infection. Consequently, infection with M. interjectum may have led to a false-positive TST.
Familiarity, availability, and low cost account for the current popularity and predominance of the TST for identification of LTBI. The QFT-GIT and T-SPOT assays offer distinct advantages over the TST. Substituting TSTs with IGRAs, as in our patient’s case, may prevent patients from receiving unwarranted treatment for LTBI based on false-positive TSTs, either due to prior BCG vaccination or infection with NTM, although no information on the effect of infection with M. interjectum and IGRA test results exists. As knowledge and accessibility of this alternative diagnostic test continue to grow, the clinical practice of how we investigate MTB infection will hopefully evolve in a fashion benefiting the patients we serve.
latent tuberculosis infection
tuberculin skin test
interferon gamma release assays
purified protein derivative
Food and Drug Administration
interferon-gamma release assays
tumor necrosis factor alpha
disease-modifying antirheumatic drugs
Centers for Disease Control and Prevention
QuantiFERON-TB Gold In-Tube
enzyme-linked immunosorbent assay
early secretory antigenic target-6
culture filtrate protein-10
enzyme-linked immunospot assay
peripheral blood mononuclear cells
mycobacteria other than tuberculosis