Tuberculin skin testing is a well-established tool to detect latent infection with Mycobacterium tuberculosis. Although it is an imperfect test, with both false positive and false negative results, it remains a very useful—indeed a critical—tool, both for epidemiologic research and the control and prevention of clinical tuberculosis. Unfortunately, tuberculin skin testing is very much underutilized for tuberculosis control in developing countries in Africa and Asia where there are serious HIV/AIDS epidemics. A major reason commonly given for not using the tuberculin skin test is that prior BCG vaccination interferes with the interpretation of a positive skin test.
An important study of the utility of tuberculin skin testing among a population in Guinea-Bissau is described in this issue of the journal,1 The authors studied 1059 family contacts of 220 smear-positive tuberculosis cases, and 921 contacts of 223 non-TB controls. They found that the tuberculin skin test performed very well in the detection of latent tuberculosis. A BCG scar increased the likelihood of a positive skin test only in persons who had household exposure to an active case of tuberculosis, not in the control households. The authors correctly interpret these data to indicate that tuberculin reactivity was due to latent M. tuberculosis infection among those with household contact. In the absence of such contact, BCG vaccination in infancy or childhood rarely confounds the interpretation of a tuberculin skin test several years later. Therefore, the widespread use of BCG among infants and children is not a general contraindication to the use of the tuberculin skin test.
A recently published study among medical house staff at 2 teaching hospitals in New York City found that graduates from medical schools outside the United States were less likely than U.S. medical school graduates to offer treatment to prevent tuberculosis to persons with positive skin tests.2 About half of those educated elsewhere believed that a positive skin test was usually due to a previous BCG vaccination.
We have recently published the results of a survey of a national sample of 300 physicians in Thailand, who were selected by a multistage random cluster sampling of those practicing in public hospitals throughout the country.3 Thailand has had a major outbreak of HIV since 1989, and tuberculosis is the most common opportunistic infection among patients with HIV in that country.4
It has been estimated that over 600,000 persons in Thailand are currently infected with HIV.5 Over 25% of adults have latent tuberculosis infections.6 Tuberculin skin testing of HIV-infected patients and prophylactic therapy is recommended officially by the Ministry of Public Health to prevent reactivation tuberculosis in those with positive skin tests.7 Despite these official recommendations, only 58 (19%) of the surveyed physicians reported that they screened HIV-infected patients for latent tuberculosis and provided therapy to skin-test reactors to prevent activation.
Many (40%) of these physicians stated that skin testing reagents were not available in their hospital. Also 25% were concerned that patients would not return to have their skin test read, and 33% felt that INH therapy for latent tuberculosis infection would induce resistance or would be too toxic. One-third of the physicians who reported treating PPD-positive patients for latent tuberculosis infection did not do chest radiographs to rule out active tuberculosis but relied on the absence of clinical symptoms before initiating therapy.
The underutilization of skin testing and treatment of those with latent TB infection to prevent reactivation of tuberculosis for skin test positives is also a problem among physicians in the United States. A study among 630 physicians in San Francisco in 1995–1996 found that only 34% were aware of the US Public Health Service guidelines for the prevention of tuberculosis among HIV-positive patients and just 39% did annual tuberculin skin testing and offered treatment of latent TB infection to skin test reactors.8
Clearly, tuberculin skin testing of HIV-positive subjects and those recently exposed to an active case of tuberculosis are not as important as early detection and treatment of active cases of tuberculosis (completion of a course of effective therapy, generally with directly observed therapy, short course, or DOTS). Nevertheless, latent tuberculosis infection also deserves detection and treatment as part of any comprehensive public health program to control tuberculosis.
It is of concern that skin testing and treatment of latent tuberculosis are rarely used in areas where they would be most useful. In some countries with an active DOTS program, the tuberculosis epidemic has continued to expand,9,10 especially in countries with a high prevalence of HIV infection. Additional strategies beyond DOTS are needed to control tuberculosis in these countries.9
There are, of course, some legitimate concerns about the implementation of a widespread program of skin testing and treatment of latent tuberculosis infection. Inadequate screening of skin-test-positives to rule out active tuberculosis could lead to monotherapy of some patients with active TB and thus promote drug resistance. Some patients will not return to have their skin test read. However, only about 15% of HIV-positive patients who were enrolled in a prophylactic trial in Thailand failed to return.11 In addition, some patients who are offered treatment of latent tuberculosis infection will fail to adhere to a course of drug therapy. Still, with adequate counseling and frequent follow-up, over 80% of patients adhered to 6–9 months of therapy for latent tuberculosis in 2 reports from Thailand.11,12 In spite of these potential pitfalls of therapy for latent tuberculosis, a recent Cochrane review of 11 trials (which included 8130 HIV-positive participants) found an overall lower incidence of active tuberculosis in treated persons, (relative risk = 0.64; 95% confidence interval = 0.51–0.81).13 The benefit was greater in persons with a positive skin test, (0.38; 0.25–0.52) than in those who had a negative skin test (0.83; 0.58–1.18). In 11 trials involving 73,375 HIV-uninfected patients the relative risk of tuberculosis was 0.40 (0.31–0.52) after 6 or 12 months of isoniazid prophylaxis.14
Highly sensitive and specific in vitro assays for latent TB infection include new diagnostic tests that detect the secretion of interferon gamma by peripheral mononuclear cells after exposure to RD1 antigens (ESAT6 and CFP10) from M. tuberculosis. These tests may be superior to traditional tuberculin skin tests for the detection of latent tuberculosis infection in some situations.15 However, these tests require a fairly sophisticated laboratory and little or no data are available regarding their use in immunocompromised persons, especially those with HIV coinfection.
For the present, tuberculosis control programs in developing countries with HIV epidemics should expand both the routine use of the tuberculin skin test to screen high-risk populations, and treatment to prevent activation of latent tuberculosis. The evidence from Guinea-Bissau suggests that prior BCG vaccination should not be an obstacle in the provision of such services.
ABOUT THE AUTHOR
KENRAD E. NELSON is Professor of Epidemiology at the Bloomberg School of Public Health, Johns Hopkins University. His research interests are the epidemiology and prevention of HIV infections and AIDS-related opportunistic infections and viral hepatitis in the United States, Southeast Asia and the Republic of Georgia. He is the editor of Infectious Diseases Epidemiology, Theory and Practice, 2nd edition.
1. Gustafson P, Lisse I, Gomes V, et al. Risk factors for positive tuberculin skin test in Guinea-Bissau. Epidemiology
2. Hirsch-Moverman Y, Tsiouris S, Salazar-Schicchi J, et al. Physician attitudes regarding latent tuberculosis infection, international vs. US medical graduates. Int J Tuberc Lung Dis
3. Hiransuthikul N, Hiransuthikul P, Nelson KE, et al. Physician adherence to isoniazid preventive therapy guidelines for HIV-infected patients in Thailand. Southeast Asian J Trop Med Public Health
4. Chariyalertsak S, Sirisanthana T, Saengwonloey O, et al. Clinical presentation and risk behaviors of AIDS patients in Thailand 1994–1998, regional variation and temporal trends. Clin Infect Dis
5. UNAIDS. Report on the Global AIDS Epidemic
. Geneva, Switzerland: WHO; 2005.
6. Siriarayapon P, Yanai H, Glynn JR, et al. The evolving epidemiology of HIV infection and tuberculosis in Northern Thailand. J Acquir Immune Defic Dis
7. Department of Communicable Disease Control, MOPH, Thailand and World Health organization. Second Review of the National Tuberculosis Program in Thailand
. Bangkok, Thailand: Ministry of Public Health; 1999.
8. DeRiemer K, Daley CL, Reingold AL. Preventing tuberculosis among HIV-infected persons: a survey of physicians knowledge and practices. Prev Med
9. DeCock KM, Chaisson RE. Will DOTS do it? A reappraisal of tuberculosis control in countries with high rates of HIV infection. Int J Tuberc Lung Dis
10. Kenyon TA, Mwasekaga MJ, Huebner R, et al. Low levels of drug resistance amidst rapidly increasing tuberculosis and human immunodeficiency virus co-epidemics in Botswana. Int J Tuberc Lung Dis
11. Hiransuthikul N, Nelson KE, Hiransuthikul P, et al. INH preventive therapy among adult HIV-infected patients in Thailand. Int J Tuberc Lung Dis
12. Ngamvithayapong J, Uthaivoravit W, Yanai H, et al. Adherence to tuberculosis preventive therapy among HIV-infected persons in Chiang Rai, Thailand. AIDS
13. Woldehanna S, Volmink J. Treatment of latent tuberculosis infection in HIV-infected persons. Cochrane Database of Syst Rev
14. Smieja MJ, Marchetti CA, Cook DJ, et al. Isoniazid for preventing tuberculosis in non-HIV infected persons. Cochrane Database of Syst Rev
15. Pai M, Riley LW, Colford JM Jr. Interferon-gamma assays in the immunodiagnosis of tuberculosis: a systematic review. Lancet Infect Dis