The April issue of Pediatrics in Reviewhas a couple of excellent articles reviewing Mycobacterium tuberculosis (MTB) and non-tuberculous mycobacterial (NTM) infections. Both these infections are often a diagnostic dilemma for pediatricians. We are frequently dependent on testing to confirm or exclude these infections.
The gold standard for the diagnosis of both of these infections is the isolation of the organism by culture. However, in many clinical circumstances, this is not feasible due to difficulties obtaining adequate samples in children (eg, gastric aspirates, sputum, tissue, etc).
We use less invasive tests as surrogate tools for the diagnosis, based on the host’s response to the infection: tuberculin skin test (TST) and Interferon gamma release assays (IGRA). However, both tests have advantages as well as their own set of limitations.
TST is the oldest test and the test that pediatricians are more familiar with. In fact, provider familiarity has been listed as one of the advantages of this test. Although providers are familiar with TST, the reality is that there is a lack of proficiency in the administration, and especially, the reading and interpretation of the test. The range of inter-reader variability is wide, and this variability is getting worse as the number of TB cases in the United States continues to decrease. How many true positive TSTs do primary care providers actually see? This is a huge disadvantage. Another significant disadvantage of TST is the cross-reaction with NTM, which is relatively common in young children.
We would add as a disadvantage the fact that there is no standard way to record and/or retrieve the test results in the medical records. Results are recorded in providers’ notes, nurses’ notes, other notes, or not recorded at all. Where do you start looking for the results? This is not ideal for patient care, infection control, or public health/epidemiology.
IGRAs have been shown to have a comparable sensitivity, but higher specificity than TST. There is little to no inter-reader variability, and the results can be easily retrieved and tracked. Although IGRAs were FDA approved in 2005, pediatricians were slow to embrace IGRAs because initially they were not recommended for all age groups and required venipuncture.
More recent data has supported changes in the recommendations on the use of IGRAs in young children. A recently published study looked at the different tests to diagnose TB infection (TST, Quantiferon Gold In-tube® Qiagen, Germantown, Maryland, and T-SPOT.TB® Oxford Immunotec, Marlborough, Massachusetts) in populations at high risk for TB in the United States, using a statistical technique known as latent class analysis. A subgroup of the population consisted of 464 foreign-born, HIV-negative children aged younger than 5 years. The estimated prevalence of latent tuberbulosis infection (LTBI) in this subgroup was 4%; the sensitivity of IGRAs and TST was comparable (59-70% vs 75%); the specificity of IGRAs was significantly higher than TST (99% vs 70%); and the positive predictive value of IGRAs was also significantly higher (71-74% vs 10%) 1. For a clinician, this means that 90% of foreign-born HIV-negative children younger than 5 years with a positive TST do NOT have LTBI, according to this study. This and other studies led to the current AAP recommendations for TB screening using IGRAs in children as young as 2 years of age, especially if they are BCG-vaccinated 2. Of note, some infectious diseases experts use IGRA in children younger than 2 years of age, and there are increasing data supporting this practice. Caution in the interpretation of these tests in very young children is recommended because they are at high risk of developing TB disease (especially severe disease) if they are exposed. In a child less than two years of age, a positive IGRA test in most cases would indicate TB infection, but a negative IGRA test would not rule out TB. If the child has been exposed to a case of TB or if TB disease is suspected, many experts recommend doing both IGRA and TST. However, it is important to remember that in these cases, neither test can definitely rule out TB.
In practical terms, immigrant children 2 years of age or older who have received BCG should be screened for TB using IGRAs. If such a child has a TST placed first and it is positive, AND the child is asymptomatic and has no risk factors for TB other than the country of origin, the child should be tested using IGRA. If the IGRA is negative, the TST is considered a false positive and the child does not need to be treated for LTBI. If IGRA is performed first, there is no need to do TST, regardless of the results, except in high-risk situations such as clinical suspicion of TB, or high risk of progression to TB disease if infected (eg, immunocompromised patients or young infants exposed to active/contagious TB cases). In these latter cases, a positive result in any of the tests should be considered a true positive. Epidemiological history in TB is as important as testing, if not more so. Even with better tests, the diagnosis of TB still requires a high index of suspicion. A supporting clinical scenario, especially suspicious radiologic findings, in a patient with risk factors should suggest to the practitioner that TB is likely, regardless of TST or IGRA results.
One last important comment: The options for LTBI treatment have changed since the review and finalization of the article on TB published in Pediatrics in Review. Nine months of INH is no longer considered the preferred regimen. Any of the regimens listed is considered adequate and equally safe. In fact, most TB experts prefer the shorter regimens due to significantly higher rates of completion, and the AAP supports this recommendation 2, 3.
References
- Stout JE, Wu Y, Ho CS, et al. Evaluating latent tuberculosis infection diagnostics using latent class analysis. Thorax 2018; 73: 1062-70
- American Academy of Pediatrics. Tuberculosis. In: Kimberlin DW, Brady MT, Jackson MA, Long SS, eds. Red Book: 2018 Report of the Committee of Infectious Diseases. 31st ed. Itaska, IL: American Academy of Pediatrics; 2018: 828-53
- Borisov AS, Bamrah Morris S, Njie GJ, et. al. Update of Recommendations for Use of Once-Weekly Isoniazid-Rifapentine Regimen to Treat Latent Mycobacterium tuberculosis Infection. MMWR Morb Mortal Wkly Rep 2018; 67(25): 723–6
