We describe clinical features and essential laboratory data in 24 patients who underwent paradoxical clinical deterioration while receiving highly active antiretroviral therapy (HAART) for the treatment of human immunodeficiency virus (HIV) infection. The patients had responded to antiretroviral therapy as defined by increased CD4+ T cell lymphocyte counts and reductions in HIV viral loads. In 14 of 24 patients, the deterioration was attributable to a response to an infectious agent proven to be present before starting HAART. While 8 of the remaining 10 patients developed inflammation against microbes that had not been clinically recognized before starting HAART, it seems likely that these organisms were already established before the institution of therapy. Two of these patients had mycobacterial disease of the colon, 2 had symptoms secondary to disseminated cytomegalovirus, 1 had lymphadenitis in association with Bartonella infection, and 1 had an atypical pneumonia in response to Pneumocystis carinii. Presumably latent infections with herpes viruses were responsible for symptoms in 2 cases; 1 patient developed severe periproctitis secondary to herpes simplex while another had progressive Kaposi sarcoma, a disease associated with human herpes virus-8. Of the final 2 patients, 1 had worsening of sarcoidosis, a disease in which CD4+ T lymphocytes play a central role in pathogenesis, while the other developed an autoimmune disorder (Graves disease), which may have been related to the production of autoreactive T lymphocytes occurring during immune reconstitution. We propose that a restored capacity of the host to mount an inflammatory response against persistent microbial antigens or self-antigens led to the development of symptoms in these patients; we call this constellation of events the immune reconstitution inflammatory syndrome (IRIS). We found reports of an additional 158 cases of IRIS. We discuss here the literature on the basic science of immune reconstitution engendered with HAART, along with microbial-specific immune responses that occur following HAART. The preponderance of cases occurred in association with Mycobacterium avium complex (28 cases), Mycobacterium tuberculosis (34 cases), cytomegalovirus (19 cases), or herpes zoster (38 cases). Other associated diseases included Cryptococcus neoformans meningitis and lymphadenitis, Kaposi sarcoma, Hepatitis B, Hepatitis C, progressive multifocal leukoencephalopathy, Graves disease, and sarcoidosis. For some presentations, such as sarcoidosis, cryptococcal meningitis, and herpes zoster, the clinical manifestations were quite similar to the same disease occurring in the absence of HAART. Other presentations, such as localized Mycobacterium avium complex lymphadenitis, cytomegalovirus vitreitis, and inflammatory PML, were unique to patients with immune reconstitution. Anecdotal reports suggest that there may be some benefit from treatment with antiinflammatory drugs if the inflammatory response compromises vital structures or has caused substantial systemic symptoms. The advent of HAART has improved the prognosis of persons infected with HIV. Clinicians involved in the care of these patients need to be aware of the potential, however, for this therapy to cause a paradoxical decline in clinical status. While the search for new infectious processes must be diligent and the possibility of drug toxicity needs to be evaluated, it is important to recognize that improved immune function, itself, can be the source. In IRIS, the need to ameliorate the signs and symptoms of immune reconstitution has to be balanced with the desire to maintain the patient on long-term, effective anti-HIV therapy. With increasing recognition of patients with IRIS, it may now be possible to initiate clinical trials to determine the most appropriate approach to treatment.
