G. Positive Tuberculin Skin Test

Introduction

The human defensive response against tuberculosis (TB) largely entails cell-mediated immunity (CMI). This is in contradistinction to the humoral or antibody-mediated response to most bacterial infections. Delayed-type hypersensitivity (DTH), a closely related but not identical phenomenon to CMI, is manifested clinically as a type IV immune response mediated by lymphocytes and is characterized by indurated response to the intradermal injection of protein from the cell wall of the tubercle bacillus. The most commonly used test employs the Mantoux technique wherein a small amount of tuberculoprotein (purified protein derivative, or PPD), is introduced into the intradermal tissues with a small-gauge needle. The amount of induration is measured between two and five days later.

Historically, the reaction was originally regarded as either “positive” or “negative”. However, recently, in an effort to maximize the utility of this diagnostic test, reactions have been categorized by different criteria depending on the circumstances of the patient. This is the so-called “5-10-15 millimeter system”. The clinical applications of this stratified interpretation system, as applied to preventive chemotherapy, are displayed in Table 13. This stratification is an effort to make the tests both relatively more sensitive and more specific. Problems with interpretation will be discussed more fully below.

Who Should Have the Tuberculin Skin Test (TST)?

Diagnosis of Disease. Historically, the TST has been regarded as a major element in the diagnosis of tuberculosis. This, however, may not be a wholly appropriate use of this test. There are three potential pitfalls to the TST in clinical care: false negatives, false positives, and true but irrelevant positives. The major problem with the TST in the diagnosis of active disease is that substantial numbers of patients with active TB do not have significant reactions to the test. In various studies among HIV-negative adult patients with active pulmonary disease, 15% to 25% of individuals at the time of diagnosis did not have a positive response. As might be expected, patients with HIV infections/AIDS have progressively higher rates of anergy or false negative reactions to the tuberculin test as their CD4-lymphocyte count diminishes. Among persons with HIV infections but no other AIDS-defining illnesses and high numbers of CD4 lymphocytes, the likelihood of a positive skin test is approximately 70%. However, among patients with advanced AIDS and CD4+ T-lymphocyte depletion, as few as 10% to 20% of the patients with tuberculosis may react to the TST. These false negatives are of potentially major consequence if clinicians allow this test to be a major element in their diagnostic thinking, with the erroneous notion that “a negative TST excludes TB.” Furthermore, the TST may yield “false positive” reactions (see below). In addition, patients may have a true-positive TST, but not have active TB, i.e., a true but irrelevant positive reaction. Hence, it is recommended that the tuberculin skin test be regarded only as a diagnostic aid without major significance in terms of ruling in or ruling out active tuberculosis.

In summary, a significant tuberculin reaction increases the likelihood of any given condition being tuberculous in etiology. But the test must be used with full recognition of its limitations. Clinicians should ask themselves, how will a positive or negative result affect their thoughts and actions?

Diagnosis of Infection. Probably the most appropriate utilization of the TST is to identify people who have latent infection with TB. Among HIV-negative individuals infected with Mycobacterium tuberculosis, authorities estimate that 10% to 15% will eventually develop active tuberculosis. The TST is used to identify individuals with latent infections so that they may be offered chemotherapy to prevent progression from clinically inapparent infection to active disease. In this setting, the TST is the only diagnostic tool. Because the test defines the condition, we are not able to tell whether false negative reactions occur. However, it is reasonable to presume that some portion of individuals harboring latent tuberculosis do not react to the TST. This is particularly the case with infants or HIV-infected persons who have been recently exposed to a case of communicable tuberculosis. Another variety of “false negative” TSTs merits attention: tests which truly yield a significant amount of induration, but are erroneously read by health professionals as negative. Repeated studies demonstrate that unpracticed physicians and nurses “read” TSTs erratically with a tendency toward underestimation.

Given the relative paucity of tuberculosis in the general population of the United States, not all persons are candidates for screening with the TST. Current epidemiology indicates that the groups shown on Table 13 are at relatively high risk of tuberculosis. These persons might be routinely screened with the TST attempting to identify latent infection and to offer preventive chemotherapy. But given the propensity of the tuberculin skin tests to generate false positive results, it may be argued that tuberculin skin tests should not be used on individuals at low risk of tuberculosis infection. This is due to the likelihood of generating significantly more false positive reactions than true positives. If the tuberculin skin test is used in individuals with a high likelihood of having underlying tuberculosis infection, it performs quite well. However, when used in a population with a tuberculosis infection prevalence of less than 5% (an accurate representation of the white indigenous population of the United States), it will predictably result in substantially more false positive reactions than true positive reactions, a situation that is not helpful in terms of individual or public health.

False positive reactions to the TST have been clearly identified in relationship to prior infections with nontuberculous mycobacteria (NTM) or to vaccination with bacille Calmette-Guérin (BCG) a vaccine prepared from an attenuated but living strain of M. bovis—an organism closely related to the tubercle bacillus. NTM infections are widely distributed across the United States, but are most common in the warm, moist environment of the Southeast. Hence, for individuals who have resided in this region, particular caution should be exercised in interpreting TST results. Also, TST results in persons who have received BCG vaccination are problematic. Most individuals who receive a BCG vaccination become only transiently (6 months to 5 years) reactive to tuberculin. If no other BCG vaccinations are given and if infection with TB or NTM does not occur, after ten to twenty years, the great majority of these persons will not react to TST. Thus, authorities have indicated that most individuals from parts of the world that are endemic for tuberculosis who have positive TSTs, should be presumed to be infected with TB. Although there will be some false positive reactions when using this model, the implications of missing latent tuberculosis infection—and the opportunity to administer preventive chemotherapy—among this high risk group are regarded as unacceptable.

Surveillance

Again, the TST is the only useful tool for surveillance of populations at risk for TB infection. In this setting, there is interest both in finding the infection in individuals, so that they may be offered preventive chemotherapy, and for the detection of recently-transmitted infection, in order to monitor the risk of transmission in an institution or community. In this setting, care must be taken to avoid false positive skin test conversions. These false positives may result in inappropriate administration of INH preventive therapy, an intervention which has modest but real risk of drug toxicity and which predictably produces great anxiety.

False negative initial test results may also pose a problem. To avoid this sort of confusion, authorities have recommended that individuals to be surveyed should have two-step testing to take advantage of the phenomenon of “boosting”. Boosting overcomes the fact that with the passage of time, TST reactivity may wane. However, the placement of the first TST, to which the reaction is negative, stimulates waned immunity and causes the next TST to become positive. If the individual is given TST #1 at the time of enrollment, and no other testing is done for several months, when TST #2 becomes positive the inference must be that the individual has been newly infected during the interval period. Instead, the newly positive test merely reflects boosted DTH. To avoid this, TST #1 and #2 should be given within one to four weeks of each other. If the first test is negative, test #2 is applied at 7 days to 21 days after the initial test. If the second test is reactive, the conclusion is that the result reflects boosting, not new infection. While the likelihood of boosting is greater with advanced years (the period of time for waning is greater in older persons), virtually any population to undergo surveillance should have two-step testing to avoid erroneous interpretations that the second positive TST represents a conversion. It should be noted that boosting may occur due to prior infection with M. tuberculosis, a prior BCG vaccination, or prior infection with NTM.

Current public health practices and OSHA guidelines indicate that healthcare workers including laboratory personnel, autopsy suite personnel, and medical and nursing students should all have routine tuberculin surveillance to detect new tuberculosis infection. This umbrella has been extended to staffs and clients of correctional facilities and nursing homes where endemic and epidemic tuberculosis has been noted to occur. Current OSHA guidelines indicate that institutions should establish a regular policy for tuberculin skin testing for all employees regardless of exposure to patients or clients. This may not be appropriate given the performance characteristics of the test; nevertheless, these guidelines are presently in effect. In these settings, annual skin testing is indicated. However, testing up to twice yearly has been advocated for persons at high risk of exposure, such as those working in inner city emergency rooms, intensive care units, and TB or AIDS wards.

What to Do for Persons Found to React to the TST

If a child or adult is found to have a significantly positive TST, several issues arise:

• Where did they become infected? Particularly, among younger persons or adults with recently converted TSTs, it is critical to attempt to locate the person(s) responsible for transmitting the infection. This is important both to identify the source case, so that they may be treated, and to determine whether others may also have been infected. The investigation should be directed to clarify whether transmission occurred at home, at work, at school, or in social settings. This function is generally undertaken by the community Public Health Department.

• What additional work-up is indicated? The critical immediate question is whether the individual with the positive TST has active TB. To investigate this possibility, clinicians should conduct a careful history and physical examination, as well as perform screening laboratory studies.

Elements of this history which merit particular attention include any systemic symptoms mindful of TB, such as feverishness, sweats, weight loss, malaise, failure to thrive or to achieve regular growth landmarks for infants or children. Also, organ-specific symptoms such as cough, phlegm, hemoptysis, chest pain, dyspnea, swollen glands, masses, difficulty with urination, hematuria and/or back/hip/joint pain should be explored.

Physical examination should include focused attention upon lymphadenopathy (particularly in the supraclavicular, anterior cervical and posterior triangle chains), paraspinal/hip/inguinal masses, adnexal masses or adhesions in women, and, among persons with AIDS, cutaneous nodules.

Laboratory screening studies should include a PA and lateral chest x-ray, a complete blood count with an erythrocyte sedimentation rate, and a urinalysis. Other studies should be obtained in accordance with findings in the history and physical examination. Sputum smears and cultures for acid-fast bacilli (AFB), are mandatory for patients with chest x-ray abnormalities or suggestive respiratory symptoms. Gastric aspirates may be obtained for infants who are unable to cooperate with sputum collection.

• What therapeutic intervention(s) is(are) indicated? If the initial data suggest that the patient does have or may have active TB, treatment should be initiated and the “suspected” case reported to local public health authorities for contact tracing. In some communities or special situations, public health systems may supervise treatment of the patient, particularly if directly-observed therapy is to be administered. If the final results of the initial evaluation indicate that the patient does not have active TB, the initial multidrug treatment may be regarded as intensive preventive chemotherapy. In these cases, treatment can be terminated at three to four months and the patient is considered to have received a full-course of preventive therapy (See American Thoracic Society/Centers for Disease Control Guidelines).

On the other hand, if there is no evidence in the initial evaluation to suggest active TB, consideration should be given to isoniazid (INH) preventive therapy. Current indications for INH chemoprophylaxis are presented in Table 7-1. Notable aspects of this document include the variable criteria for “significant” tuberculin reactivity, differing recommendations for the duration of INH therapy (twelve months for HIV-infected persons or persons with inactive apical fibrotic lesions, nine months for infants or children, and six months for all others) and practices to monitor for potential drug toxicity.

When to Refer

As noted above, all cases of proven or suspected TB should be reported to local public health authorities. This does not mean that management of the patient will necessarily be relinquished, although the option of directly-observed therapy should be explored.

Patients who require bronchoscopy, biopsies, or aspirations of organs, nodes or masses should be sent to appropriate specialists.

Medicolegal Concerns

Probably the greatest areas for concern are failure to recommend isoniazid preventive therapy for persons with a positive TST who are at high risk for TB, e.g., an HIV-infected individual, and, once chemoprophylaxis is begun, failure to monitor adequately for INH-associated hepatitis.

References

American Thoracic Society. Treatment of tuberculosis and tuberculosis infection in adults and children. Am J Respir Crit Care Med 1994;149:1359-1374. These guidelines were published in 1994 and describe both treatment of active disease and the preventive therapy of persons with latent infection. The policies were also endorsed by the American Academy of Pediatricians.

American Thoracic Society. Control of tuberculosis in the United States. Am Rev Respir Dis 1992;146:1623-1633. This statement provides detailed information regarding diagnostic, screening practices, institutional infection control and performing TSTs.

Kendig EL Jr., Kirkpatrick BV, Carter WH, Hill FA, Caldwell K, Entwistle M. Underreading of the tuberculin skin test reaction. Chest 1998;113:1175-1177. A recent survey in which practicing pediatricians, academic faculty, residents and nurses read a known positive TST in comparison with trained TST readers. The group grossly underestimated the extent of induration with one-third of the interpretations being falsely negative.

Menzies R, Vissandjee B, Amyot D. Factors associated with tuberculin reactivity among the foreign-born in Montreal. Am Rev Respir Dis 1992;146:752-756. Since the foreign-born now comprise roughly 40% of new tuberculosis cases annually in the United States, screening immigrants for potential INH preventive therapy is recommended. This report highlights the yield of such screening and reviews the problem of “boosting” secondary to a prior BCG vaccination.

F. Wheezing-Stridor < back | next > H. Pleural Effusion
10 of 23 HTML pages