A test of tuberculin quality: tried and true or tired and tattered?

It is a sad state of affairs when international efforts to control the number-one infectious killer worldwide must rely heavily on a suboptimal diagnostic test developed over 100 years ago and improved upon minimally in the subsequent century. In this month's Journal, Dr. Rangel-Frausto and colleagues add to accumulating evidence reinforcing the pressing need to develop a workable alternative to tuberculin skin testing for the diagnosis of tuberculosis infection. Robert Koch announced the discovery of tuberculin in 1890. Initially touted as a potential therapeutic agent, it was quickly developed as a diagnostic tool by several methods. The Mantoux intracutaneous test remains in use today. As we enter the new millenium, tuberculin skin testing remains the only practical method for screening populations for asymptomatic infection by tuberculosis. Mantoux skin testing is frequently cited as being only 75% to 90% sensitive for active tuberculosis disease. The long list of potential causes of false-negative skin-test reactions includes a variety of infections (among them, overwhelming tuberculosis), metabolic and nutritional deficiencies, immunosuppressive drugs and immunodeficiency diseases (notably acquired immunodeficiency disease). The specificity of the test is more difficult to estimate but probably is less than 95%, with infection by nontuberculous mycobacteria and vaccination with bacille Calmette-Guerin (BCG) the most common causes of false-positive reactions. Compounded by the vagaries of inconsistent placement and interpretation, differences in quality between batches and brands of product, and the effects of concurrent medical conditions on the sensitivity and specificity of the test, the predictive value of a result in many populations is limited at best. In this issue of the Journal, Rangel-Frausto and colleagues highlight the importance of having quality product available to maximize the utility of the tuberculin test. They note that the potency of tuberculin supplied from the World Health Organization to Argentina and subsequently prepared and used in Mexico was significantly lower than advertised. The unpredictability of the test is compounded by a system of production, quality control, and administration that may involve numerous different countries and assumes that tests in guinea pigs reliably predict results in humans. There have been several reports of false-positive tuberculin skin tests in the United States as well, many of which have been attributed to problems with a particular brand or batch of tuberculin. Even the rigorous US Food and Drug Administration standards do not guarantee uncomplicated use of the product.