Psychological influences on immunity. Implications for AIDS.

" There is considerable variability in the clinical course of individuals infected with the human immunodeficiency virus (HIV), the acquired immune deficiency syndrome (AIDS) virus. Because there is good evidence for psychological mediation of immune function, psychosocial or behavioral variables are among the possible cofactors that may influence HIV infection and disease progression. This article reviews relevant psychoimmunology research and addresses the implications of these data for the lives and medical treatment of HIV-infected people. The human immunodeficiency virus (HIV) is the etiologic agent for the acquired immunodeficiency syndrome (AIDS). The extreme suppression of the immune system that is characteristic of AIDS leaves individuals vulnerable to "opportunistic" infections, diseases that are generally not a threat to healthy individuals. However, exposure to HIV does not necessarily result in seroconversion (i.e., immunologic changes reflecting HIV infection), moreover, HIV-infected individuals do not necessarily exhibit any clinical symptoms. The reasons for the progression from an asymptomatic seropositive status to the development of AiDS-related complex (ARC) are not well understood; similarly, only a certain percentage of ARC patients thus far have developed AIDS, and the time between development of AIDS and death is highly variable (Solomon & Temoshol¢, 1987). Physical cofactors that appear to promote seroconversion and disease progression include drug use, repeated exposure to HIV, other concurrent viral infections, and poor nutrition (Kaplan, Johnson, Bailey, & Simon, 1987). Because studies with both humans and animals have provided good evidence that behavior can influence immune function, psychosocial or behavioral variables are additional potential cofactors for HIV infection and disease progression. In this article we summarize relevant literature addressing psychological influences on immunity and discuss possible implications of this work for the lives and medical treatment of HIV-infected individuals. Background: Basic Immunological Concepts In order to measure immune function, the numbers and functional abilities of subgroups of lymphocytes (white blood cells) are assessed. Different lymphocyte subpopulations perform specialized functions, and there is no single immunological assay that provides a global measure of immune system competence. Because of the interdependence of various immunological components, adverse changes in one lymphocyte subpopulation can produce cascading effects. Behavioral influences on immunity are thought to be mediated in part through the endocrine system. The endocrine system is responsive to a variety of emotional states (Baum, Grunberg, & Singer, 1982), and there is good evidence for endocrine and neuroendocrine modulation of immunity (Ader, 1981). Unlike many autonomic and hormonal changes that can occur within seconds or minutes, most of the immunological components we will discuss take days or weeks to change significantly. Although some biochemical mediators may be synthesized in hours, significant changes in most lymphocyte subpopulations take considerably longer amounts of time. Therefore, a bad afternoon is probably not a sufficient stimulus to produce immunological changes of importance; however, several days of heightened dysphoria can alter a number of immunological parameters. Several immunological terms will be used in this article. The cellular immune response, one of the immune system's two arms, is particularly important for the defense against viruses such as HIV; cellular immunity refers to immune functions that primarily involve T-lymphocytes. T-lymphocytes, derived from the thymus, have a number of subgroups with important functions; for example, T-lymphocyte subclasses synthesize lymphokines, such as gamma interferon and the interleukins that function as communication links between immune and nonimmune cells and thus serve as potent immunological mediators. Helper T-lymphocytes stimulate the production of a number of other immunological activities, particularly the production of antibodies by B-lymphocytes. Suppressor T-lymphocytes act in a feedback loop to shut off the activities of helper T-lymphocytes when sufficient antibody has been produced to ward off a specific infection. HIV infects and ultimately destroys helper T-lymphocytes, disabling a very important immunological function. Whereas there are normally about twice as many helper T-lymphocytes as suppressor T-lymphocytes, excessively low helper-suppressor ratios are characteristic of AIDS because of the depletion of helper cells. For blastogenesis, a common assay, lymphocytes are 892 November 1988 • American Psychologist Copyright 1988 by the American I~chok~ical Association, Inc. 0003-066X/88/$00.75 Vol, 43, No. I 1,892-898 cultured for several days with a mitogen, a substance that stimulates cell growth and cell division. The assay is thought to provide a model for the body's lymphocyte proliferative response to foreign substances such as viruses or bacteria. Blastogenesis is clearly depressed among AIDS patients (Schechter et al., 1987). Natural killer (NK) cells are thought to provide an important defense against virus-infected cells and cancer cells (Bloom, 1980), and the ability of NK cells to destroy infected cells is clearly impaired in AIDS (Schechter et al., 1987). Interferon, a lymphokine, is a potent enhancer of various immune functions, including NK cell activity. Interferon has been of considerable interest as an antiviral agent; it also inhibits tumor growth in several tumor systems (Bloom, 1980). Immunological Changes Associated with Commonplace, Transient Events Commonplace stressful events can have significant immunological consequences. In a series of studies, it has been shown that immunological changes occur in medical students' blood samples taken during examinations compared with "baseline" samples taken one month previously when the students were not taking examinations. Examination blood samples had lower NK cell activity than samples obtained one month earlier (Glaser, Rice, Speicher, Stout, & Kiecolt-Glaser, 1986); in addition, lonelier students had lower levels of NK cell activity than their less lonely colleagues (Kiecolt-Glaser, Garner, Speicher, Penn, & Glaser, 1984). Gamma interferon showed precipitous, cyclical decrements (i.e., values during examination periods that were 5% or less of baseline values) in students followed across an academic year (Glaser et al., 1987). Blastogenesis was also lower during examinations (Glaser, Kiecolt-Glaser, Stout et al., 1985). Although not a consistent finding, examination stress has been associated with decrements in helper T-lymphocytes as well (Glaser, Kiecolt-Glaser, Stout et al., 1985; KiecoltGlaser et al., 1986). The data from the medical student studies are important because medical students have long histories of successful test-taking behavior. In spite of their familiarity with this stressor, they still show reliable affective changes (i.e., greater distress during examinations), as well as concomitant immunological changes. These data suggest that even very commonplace or frequently experienced stressors can affect immune function. Interpersonal Relationships and Immunity There is good evidence that interpersonal relationships have health-related consequences. Some of the most persuasive evidence comes from prospective epidemiological studies that show greater morbidity and mortality in peoWork on this article was supported in part by Grants R01 MH42096 and R01 MH40787 from the National Institute of Mental Health. Correspondence concerning this article should be addressed to Janice K. Kiecolt-Glaser, Department of Psychiatry, Ohio State University College of Medicine, 473 W. 12th Ave., Columbus, OH 43210. pie with fewer close relationships (Cohen & Syme, 1985). Epidemiological and immunological data suggest that both the quality of relationships and their disruption are important (Bloom, Asher, & White, 1978; Renne, 1971). Separatlos o r Divorce Marital disruption, either through divorce or death, appears to be one of the most stressful of life events (Bloom ct al., 1978; Verbrugge, 1979). Both bereavement and divorce are associated with very high rates of physical and emotional disorders; marital disruption is the single most powerful sociodemographic predictor of physical and emotional illness (Somers, 1979). Although most of the epidemiological literature on marital disruption does not separate causes of morbidity or mortality, studies that have done so have shown higher rates of both infectious disease and cancer (Ernster, Sacks, Selvin, & Petrakis, 1979; Lynch, 1977; Somers, 1979). On the basis of these epidemiological data, a crosssectional study was designed to explore the possibility of immunological changes associated with adaptation to separation and divorce. Those women who had been separated one year or less had poorer immune function (across five of six assays) than their well-matched married community counterparts (Kiecolt-Glaser, Fisher et al., 1987). Moreover, among separated and divorced women, both shorter separation periods and greater continued attachment or preoccupation with the (ex)husband were associated with poorer immune function and greater depression and loneliness. Similar data were obtained in a study with men (Kiecolt-Glaser et al., 1988).