The Problems of Diagnosing Tuberculosis

The oldest test still in use.

Question: What's wrong with the tuberculosis skin test?
 
Neil Schluger: Well, I- tuberculosis skin test is literally again one of the oldest tests in medicine. It was described in a scientific paper that was published in 1908 and is basically used in almost the same form for the last hundred years, and we've learned over a period of time that it works reasonably well but is far from perfect so it gives lots of false positive and false negative results. That's the major problem with it. In addition, it's inconvenient. Patients have to come back two days after it's placed to have the results read. The people reading the results sometimes are inaccurate so you have those problems. As I mentioned earlier, people who have had the TB vaccine can get false positive results. So there are lots of problems with it.
Topic: How it works.
Neil Schluger: So the tuberculosis skin test works in the following way. You get an injection that contains proteins that are made by tuberculosis and if your body has previously been infected with tuberculosis your immune system will remember in a sense what tuberculosis looks like, and so if you inject these proteins that come from tuberculosis under the skin your immune system as I said will remember that and cause an immune reaction, some swelling and redness at the site of the infection, and that's how the TB skin test works. In the last several years we've come to understand at a very sophisticated level exactly how the immune system recognizes those tuberculosis proteins and exactly what's going on there, the sort of molecules that are being secreted and the immune cells that are involved. And scientists working in a variety of labs realized that we could essentially recreate that skin test in a test tube and not only could we recreate it in a test tube, which makes it easier for the patient--they don't have to come back; they just give a blood sample--we could refine very much or scientists could refine very much the proteins that we used so that now we can use proteins that only represent tuberculosis and not other kinds of infections that are common, and so the problem of false positive results should be much easier to deal with. And so scientists worked on this. A couple of companies developed commercial applications of this test that are marketed now and we and many other groups have worked to see in fact how well these things do in actual practice. And I think most of us think this is going to be a real advance and will help us more accurately identify people who have been infected with TB previously so that we can treat them and not worry about treating people who just have false positive tests but really focus on the people who are at risk for getting TB and have a much more precise tool to work with.
Recorded on: 04/25/2008