On the surface, it seems reasonable to advise physicians to suspect early human immunodeficiency virus (HIV) infection when clinically indicated and to be prepared to intervene with antiretroviral therapy. However, this premise is fraught with pitfalls and challenges for physicians on the front lines of care.

Exactly who should be tested, and when? If physicians only test the highest-risk patients, the majority of patients with early HIV infection will be missed. Risk factors do not reliably discriminate between those who might be infected and those who are not infected.

A single episode of unprotected sexual intercourse might transmit HIV infection—just ask all of the young adults who have become infected without having multiple sexual partners. In fact, the highest rates of new infections are in patients who do not fall into easily defined high-risk groups. Almost any person who has been sexually active is at risk. Thus, when confronted with a person in whom early HIV infection is a possible diagnosis, the burden appears to rest on the clinician to prove that HIV is not the cause of the patient's symptoms—which brings up the next dilemma.

The symptoms of early HIV infection mimic those of many common benign viral illnesses,¹ and it is not possible to reliably exclude HIV infection as a possible cause on clinical grounds alone. There is no practical way to routinely and confidently exclude early HIV infection as a cause of common viral syndromes.

To put things into perspective: let us say that a physician practices in a college student health clinic. During the week, the physician sees dozens of students with infectious syndromes, ranging from the common cold, to influenza, to mononucleosis, to nonspecific viral illness. Many of these students will be sexually active, which makes them at risk for HIV infection. (Just think of all of the other sexually transmitted diseases that are routinely seen in a college clinic—chlamydia, herpes, trichomonas, papillomavirus, and so forth.)

So, what is the physician to do? If no such student is tested for early HIV infection, cases will be missed. On the other hand, routinely testing all students also presents many difficulties. First, because it is too early for HIV antibody to form, costly viral RNA testing must be done. Then, there is the anxiety created for students awaiting their test results. Imagine walking into a physician's office for what you thought was a bad cold and learning that you may actually have HIV infection! Students coming in for cold symptoms would have to receive an extended counseling session before testing, and many would need a second appointment to deliver the results. Multiplying that scenario many times each year could easily overwhelm an office's resources.

Is this scenario plausible? Absolutely. In a recent study, researchers evaluated serum samples for evidence of acute HIV infection from patients who were being tested for infectious mononucleosis. Approximately 1 percent of the patients in this study had unrecognized acute HIV infection.² Considering this degree of risk of a fatal disease, the argument could be made that most sexually active patients who are tested for mononucleosis should be tested for acute HIV infection if they are heterophile-negative. This strategy is not yet standard practice today.

However, physicians should gradually become more liberal in testing for early HIV infection. Currently, we do so somewhat arbitrarily, by making educated guesses rather than by being systematic. However, we should move toward testing more routinely. We must develop rapid test kits that will give readings in the office and work to overcome the technical and financial challenges that prevent us from optimizing HIV diagnosis and treatment. As discussed in the article by Perlmutter and colleagues,³ the goal is for clinical practice to keep up with the many advances in diagnosing and treating early HIV infection.