- Email: [email protected]
Epstein-barr virus antibodies and severity of depression
Correspondence
Response To the Editor: We appreciate Dr. Kramer's comments and while we agree with many of his general assertions, we feel that he has mixed up several related but distinct issues: (1) change in positive and negative symptoms with neurolepfic treatment in the acute phase of the illness (this was the question addressed in our present study) (2) the measurement of negative symptoms and their delineation from phenomena such as depression and EPS; (3) the relationship of cholinergic hyperactivity to negative symptoms and the possible role of anticholinergics in their treatment. In our study, we observed that negative symptoms improved in tandem with positive symptoms. Factors such as EPS and akinetic depression did not confound this observation as baseline negative symptom ratings (from which improvement was measured) were performed when patients were drag-free for a minimum of 2 weeks and both psychotropic-n~ve and previously treated patients showed this same pattern of improvement. Despite the phenomenological overlap between depression and negative symptoms, recent studies (including our own, Liberzon et al 1990), indicate that these phenomena can be distinctly measured. With regard to positive symptoms, most experts consider positive and negative to be distinct symptom dimensions, which may be related at some phases of the illness. Our finding do sugges! Lhat psychotic-phasic negative symptoms may differ from deficit-enduring negative symptoms with regard to neuroleptic-responsiveness and covariance wltb positive symptoms, but to further infer that the former are not "true negative symptoms" would be premature (Andreasen 1999; Tandon and Greden 1990). The implications of our findings are less clear.
Epstein-Barr Virus Antibodies and Severity of Depreasion To the Editor: While laboratory evidence of impaired cellular immune function has been reported in patients with affective disorders (Kronfol et al 1983, 1985; Schleifer et al 1984; Syvalahti et al 1985; Albrecht et al
B ~ . PSYCHIATRY 1991;29:618-625
62 l
Several mechanisms (including concomitant ~ c tion in dopaminergic and cholinergic activity) ~ be invoked to explain these findings. Thus, although these findings would be consistent with ~ ~ I of concomitant increases in dopaminergic ~ cholmergic activity in the psychotic ~ of schizophrenic illness (related to positive ~ negative symptoms, respectively), other explanatiom are plausible as well. We would like to underscore the point that ~ dopaminergic/cholinergic model ?nd the implicati~ of cholinergic hyperactivity in the productionof negative symptoms are hypotheses to be tested, ~ the study suggested by Dr. K.,mner might be one way to test it. Rajiv Tandon Robert S. Goldman John F. Greden Schizophrenia Program University of Michigan Ann Arbor, MI 48109
References Andreasen NC (1990): Foreword. In Greden IF and Tandon R (eds), Negative Schizophrenic Symptoms: Pathophysiology and Clinical Implications. Washington DC, American Psychiatric Press. Liberzon l, Goldman R, Tandon R (1990): Assessment of depression in schizopba-enia. New Research abstracts, :anencan Psychiatric Association annual meeting, New York. Tandon R, Greden IF (1990): In conclusion: Is integration possible? In Greden IF, Tan-don R (eds), Negative Schizophrenic Symptoms: Pathophysiology and Clinical Irap!ica~or~. ~ ~ o n DC, Ameficar. Psycla~amc Press.
1985; Irwin and Gillin 1987), this has not been linked to an increased risk of clinical infection in these patients. Antibodies reactive with Epstein-Ban" Viru.': (EBV)-associated antigens are present in most adults, indicating endemic latent infection with this pathogen. Elevated titers of these antibodies are seen in a variety of ~ u n o c o m p r o m i s e d states, presumably indicating persistent active or reactivated infection as
622
Correspondence
BtOL PSYCHIATRY 1991 ;~:618-625
a consequence ot deficient cell-mediated immunity (Merino et al 1986; Rinaldo et al 1986; ThorleyLawson 1988). Thus, EBV antibody titers may arguably be viewed as a crude but clinically relevant measure of cellular immunocompetence. We now report a suggestive correlation between EBV antibody titers and severity of psychiatric symptoms in a sample of b.ospitalized depressed patients. These findings complement other reports linking degree of immune dysfunction with severity of depression (Allen and Tilkian 1986; Denney et al 1988; Schleifer et al 1989), and may indicate that the immune changes associate! with depression are clinically important. EBV antibody titers were routinely obtained in depressed patients referred for admission to a psychiatric investigational unit over a 2-year period, because of reports of a putative chronic EBV syndrome resembling depiession (Salit 1985; Pitts et al 1987; Cooke 1991). Serum for antibody testing was obtained between 8:00 and 9:00 AM the first working day after admission. All patients were systematically evaluated for DSM-HI diagnoses of major affecfive disorders (American Psychiatric Association 1980), and all had standardized assessment by the Montgomery-Asberg depression rating scale (MADRS) (Montgomery and Asberg 1979) carried out between 8:30 and 10:00 AM within 3 working days of admission, as part of the ward routine assessment. Although designed to measure change in severity of depression over time, the MADRS is also a useful cross-sectional measure of severity (Keams et al 1982). Psychotropic medication use in the week leading up to admission was "alsorecorded, but reliable documentation of earlier medication use was not always available. Antibodies reactive with EBV-associated viral capsid antigen (VCA) and early antigen (EA) were detected in serum as previously described (Swanston et al 1986) in dilutions from 1:40 to 1:640 for VCA, and from 1:20 to 1:160 for EA. Statistical tests, performed on a DEC 2020 computer using the software package SPSS-X, included analysis of variance (ANOVA) (age comparisons only), Kruskal-Wallis oneway ANOVA, and the Spearman correlation test. Variables were VCA and EA titers, presence or absence of EA antibodies, polarity (i.e., major depression versus bipolar disorder, depressed phase), and medication use or not in the p~ceding week, as well as age and gender, which have been reported to influence EBV antibody titers (Sumaya et al 1975; Glaser et al 1985; King et al 1985). Preliminary data on EBV antibody prevalence in some of these patients have been reported elsewhere (Cooke et al 1988).
After exclusion of 5 subjects with undetectable VCA antibodies (indicating no prior exposure to EBV), the sample comprised 30 subjects with diagnoses of major depression (9 men, 13 women) or bipolar disorder, depressed phase (5 men, 3 women). None had evidence of acute medical illness on admission. Five of the 30 subjects had been drug free for ~ 1 week. The mean age _ S D was 40.8 - 13.0 years, and the mean MADRS score _+ SD was 27.3 ± 10.0. Neither MADR$ scores, nor antibody titers correlated significantly with age. Mean ages, MADR$ scores, and VCA and EA titers were not significantly affected by gender, polarity, or expostae to medication t,v > 0.1 for ~l~ ~ y s e s ) . MADRS scores were significantly correlated with EA titers in both sexes (males: r = 0.627, p < 0.01; females: r = 0.562, p < 0.025, see Figure 1). MADRS scores were also higher in those subjects with detectable EA antibodies than in those without (F = 6.26; df = 1,28, p < 0.025), with a significant difference seen only in males when the sexes were considered separately (F = 6.61; df = 1,12;p < 0.025). There was no significant correlation between MADRS scores and VCA titers (p > 0.1). The relationship between MADRS scores and EA antibody titers observed in our newly admitted depressed patients may reflect EBV reactivation occurring in more severely depressed subjects. If further work establishes that this is the case, then this may offer a useful clinical model for exploring the im50-
40
o o
g
8 @
i.H tr
o 30 o o0 oo rr
$
o
oo
o 20
10
0
o
I <20
I
I 40
l
i 160
Reciprocal EA Titre
Figure 1: MADRS depression scores as a function of EA antibody tite~ (shown as reciprocal value) in ~,nale (o) and female (e) depressed patients. See text for results of Spearman correlation.
Correspondence
am~. PSYClIIATRY 199| ;29:611~-625
munologic consequences of depression. We have diseussed these issues at length elsewhere (Cooke et al 1989). In our sample, exposure to medication in the week prior to admission appeared not to affect EBV tilers, but our naturalistic and uncontrolled design did not allow us to rule out possible effects of age, medical condition, or earlier medication use in our drug-free subjects. To our knowledge, the effects of psychotropic medication on EBV antibody titers have not been reported. Future studies examining the relationship between EBV titers and depression will need to have carefully matched controls. Robert G. Cooke a'b Jerry J. Warsh a'b'c.e Gary M. Hase~)a'b Bernadette J. M. McLaughlin d'f Thecla Jornae "b
~Clarke Institute of Psychiatry 250 College Street Toronto, Ontario M5T IR8. Departments of bpsychiatry, ~Pharmacology and dMicrobiology, and qnstitute of Medical Science, University of Toronto. fLaboratory Services Branch, Ontario Ministry of Health, Toronto, Ontario.
References Albrecht J, Helderman JH, Schlesser MA, Rush AJ (1985): A controlled study of cellular inmmne function in affective disorders before and during somatic therapy. Psychiatry Res 15:185-193. Allen AD, Tilkian SM (1986): Depression correlated with cellular immunity in Systemic immunodeficient EpsteinBan"Virus Syndrome.,(SIDES). J Clin Psychiatry 47:133. American Psychiatric Association (1980): Diagnostic and Statistical Manual of Mental Disorders, Third Edition, (DSM HI). Washington, DC: APA, pp 210-218. Cooke RG, The psychiatrist and c.Vaonic fatigue syndrome. (1991): Can Dis Wkly Rep (supplement) (in press). Cooke RG, Langlet F, McLaughlin BJM (1988): Age-Specific prevalence of Epstein-Barr Virus antibodies in adult patients with affective&sorders.J Clin Psychiatry 49:361. Cooke RG, Warsh J J, Hasey GM ~,1989): Epstein-Barr Virus as a cause of autoimmtme disease and other medical morbidity in patients with affective disorders. Med Hypotheses 29:177.
623
Denney DR, Stephenson LA, Pvnick EC, W e l ~ RA ( 1 ~ ) : Lymphocyte subclasses and d e ~ . J Abnoem Psychol 94:499-502. Glaser R, Strain EC, Tart KL, Holliday/E, RL, Kiecolt-Glaser IlK (1985): Changes in Epstein-Barr virus antibody titers associated with aging (421~). Proc Soc Exp Bill Med 179:352. Irwin M, Gillin JC (1987): Impaired rialto'at killer cell activity among depressed patients. P~chiawy Res 20: t 8 i182. Keams NP, Cmickshank CA, McGuig-an KI, Riley SA, Shaw SP, Snaith RP (1982): A comparison of depression rating scales. Br J Psychiatry 141:45. King DJ, Cooper SI, Earle JAP, et at (1985): A survey of serum antibodies to eight common ~ s in psychiatric patients. Br J Psychiatry 147:137. Kronfol Z, Silva J Jr, Greden J, Dembinski S, Gardner R, Carroll B (1983): Impaired lymphocyte ~ d c a in depressive illness. Life Sci 33:241-247. Krordol Z, NasF,dlah HA, Chapman S, House ]D (1985): Depression, cortisol metabolism ~ d iymphocytopenia. d Affective Disord 9:16o- 173. Merino F, Henle W, Ramirez-Ihtque P (1986): Chronic active Epstein-Barr virus infection in patients with Chediak-Higashi syndrome. J Clin lmmunol 6:299-305. Montgomery SA, Asberg M (1979): A new depression scale designed to be sensitive to change. Br J Psychiatry 134:382. Pitts FN, Allen RE, H a r ~ ~/~, Alien AD (1987): Immunodeficiency t3 ia~,-pesviruses causes depression. Paper presented at me l,~'lh AnvxmIMeeting of the American Psychiatric Association, Chicago, May 9-14. Rinaldo CR Jr, Kingsley LA, Lyter DW, et al (1986): Association of ttTLV-III vA~ Epstein-Barr virus infection and abnormalities ofT lymphocytes in homosexual men. J Infect Dis 154:556. Salit ~ (1985): Sporadic postinfectious neuromyasthenia. Can Med Assoc J 133:659. Schle:~er SJ, Keller SE, Meyerson AT, Raskm MJ; Davis KL, Stein M (1984): Lymphocyte function in major depressive disorder. Arch Gen Psychiatry 41:4~ ~86. Schleifer SJ, Keller SE, Bond RN, Cohen J, Stein M (1989): Major depressive disorder and immunity. Role of age, sex, severity, and hospitaliTation. Arch Gen Psychiatry 46:81-87. Sumaya CV, Henle W, Henle G, Smith MHD, LeBlanc D (1975): Seroepidemiologic study of Epstein-Ban" infections in a rural community, J Infect Dis 131:403. Swanston W, MahonyJ, McLaughlin B, Chemesky M (1986): Assessment of serologic markers for Epstein-Barr virus. Diagn Microbiol Infect Dis 5:235. Syvalahti E, Eskola J, Ruuskanen O, Teijo L (1985): Nonsupp~ssion of cortisol in depression and immune function. Prog Neuropsychophannacol Biol Psychiatry 9:413422 Thorley-Lawson DA (1988): Basic virological aspects of Epstein-Barr virus infection. Semin Hematol 25:247260.
Response To the Editor: We appreciate Dr. Kramer's comments and while we agree with many of his general assertions, we feel that he has mixed up several related but distinct issues: (1) change in positive and negative symptoms with neurolepfic treatment in the acute phase of the illness (this was the question addressed in our present study) (2) the measurement of negative symptoms and their delineation from phenomena such as depression and EPS; (3) the relationship of cholinergic hyperactivity to negative symptoms and the possible role of anticholinergics in their treatment. In our study, we observed that negative symptoms improved in tandem with positive symptoms. Factors such as EPS and akinetic depression did not confound this observation as baseline negative symptom ratings (from which improvement was measured) were performed when patients were drag-free for a minimum of 2 weeks and both psychotropic-n~ve and previously treated patients showed this same pattern of improvement. Despite the phenomenological overlap between depression and negative symptoms, recent studies (including our own, Liberzon et al 1990), indicate that these phenomena can be distinctly measured. With regard to positive symptoms, most experts consider positive and negative to be distinct symptom dimensions, which may be related at some phases of the illness. Our finding do sugges! Lhat psychotic-phasic negative symptoms may differ from deficit-enduring negative symptoms with regard to neuroleptic-responsiveness and covariance wltb positive symptoms, but to further infer that the former are not "true negative symptoms" would be premature (Andreasen 1999; Tandon and Greden 1990). The implications of our findings are less clear.
Epstein-Barr Virus Antibodies and Severity of Depreasion To the Editor: While laboratory evidence of impaired cellular immune function has been reported in patients with affective disorders (Kronfol et al 1983, 1985; Schleifer et al 1984; Syvalahti et al 1985; Albrecht et al
B ~ . PSYCHIATRY 1991;29:618-625
62 l
Several mechanisms (including concomitant ~ c tion in dopaminergic and cholinergic activity) ~ be invoked to explain these findings. Thus, although these findings would be consistent with ~ ~ I of concomitant increases in dopaminergic ~ cholmergic activity in the psychotic ~ of schizophrenic illness (related to positive ~ negative symptoms, respectively), other explanatiom are plausible as well. We would like to underscore the point that ~ dopaminergic/cholinergic model ?nd the implicati~ of cholinergic hyperactivity in the productionof negative symptoms are hypotheses to be tested, ~ the study suggested by Dr. K.,mner might be one way to test it. Rajiv Tandon Robert S. Goldman John F. Greden Schizophrenia Program University of Michigan Ann Arbor, MI 48109
References Andreasen NC (1990): Foreword. In Greden IF and Tandon R (eds), Negative Schizophrenic Symptoms: Pathophysiology and Clinical Implications. Washington DC, American Psychiatric Press. Liberzon l, Goldman R, Tandon R (1990): Assessment of depression in schizopba-enia. New Research abstracts, :anencan Psychiatric Association annual meeting, New York. Tandon R, Greden IF (1990): In conclusion: Is integration possible? In Greden IF, Tan-don R (eds), Negative Schizophrenic Symptoms: Pathophysiology and Clinical Irap!ica~or~. ~ ~ o n DC, Ameficar. Psycla~amc Press.
1985; Irwin and Gillin 1987), this has not been linked to an increased risk of clinical infection in these patients. Antibodies reactive with Epstein-Ban" Viru.': (EBV)-associated antigens are present in most adults, indicating endemic latent infection with this pathogen. Elevated titers of these antibodies are seen in a variety of ~ u n o c o m p r o m i s e d states, presumably indicating persistent active or reactivated infection as
622
Correspondence
BtOL PSYCHIATRY 1991 ;~:618-625
a consequence ot deficient cell-mediated immunity (Merino et al 1986; Rinaldo et al 1986; ThorleyLawson 1988). Thus, EBV antibody titers may arguably be viewed as a crude but clinically relevant measure of cellular immunocompetence. We now report a suggestive correlation between EBV antibody titers and severity of psychiatric symptoms in a sample of b.ospitalized depressed patients. These findings complement other reports linking degree of immune dysfunction with severity of depression (Allen and Tilkian 1986; Denney et al 1988; Schleifer et al 1989), and may indicate that the immune changes associate! with depression are clinically important. EBV antibody titers were routinely obtained in depressed patients referred for admission to a psychiatric investigational unit over a 2-year period, because of reports of a putative chronic EBV syndrome resembling depiession (Salit 1985; Pitts et al 1987; Cooke 1991). Serum for antibody testing was obtained between 8:00 and 9:00 AM the first working day after admission. All patients were systematically evaluated for DSM-HI diagnoses of major affecfive disorders (American Psychiatric Association 1980), and all had standardized assessment by the Montgomery-Asberg depression rating scale (MADRS) (Montgomery and Asberg 1979) carried out between 8:30 and 10:00 AM within 3 working days of admission, as part of the ward routine assessment. Although designed to measure change in severity of depression over time, the MADRS is also a useful cross-sectional measure of severity (Keams et al 1982). Psychotropic medication use in the week leading up to admission was "alsorecorded, but reliable documentation of earlier medication use was not always available. Antibodies reactive with EBV-associated viral capsid antigen (VCA) and early antigen (EA) were detected in serum as previously described (Swanston et al 1986) in dilutions from 1:40 to 1:640 for VCA, and from 1:20 to 1:160 for EA. Statistical tests, performed on a DEC 2020 computer using the software package SPSS-X, included analysis of variance (ANOVA) (age comparisons only), Kruskal-Wallis oneway ANOVA, and the Spearman correlation test. Variables were VCA and EA titers, presence or absence of EA antibodies, polarity (i.e., major depression versus bipolar disorder, depressed phase), and medication use or not in the p~ceding week, as well as age and gender, which have been reported to influence EBV antibody titers (Sumaya et al 1975; Glaser et al 1985; King et al 1985). Preliminary data on EBV antibody prevalence in some of these patients have been reported elsewhere (Cooke et al 1988).
After exclusion of 5 subjects with undetectable VCA antibodies (indicating no prior exposure to EBV), the sample comprised 30 subjects with diagnoses of major depression (9 men, 13 women) or bipolar disorder, depressed phase (5 men, 3 women). None had evidence of acute medical illness on admission. Five of the 30 subjects had been drug free for ~ 1 week. The mean age _ S D was 40.8 - 13.0 years, and the mean MADRS score _+ SD was 27.3 ± 10.0. Neither MADR$ scores, nor antibody titers correlated significantly with age. Mean ages, MADR$ scores, and VCA and EA titers were not significantly affected by gender, polarity, or expostae to medication t,v > 0.1 for ~l~ ~ y s e s ) . MADRS scores were significantly correlated with EA titers in both sexes (males: r = 0.627, p < 0.01; females: r = 0.562, p < 0.025, see Figure 1). MADRS scores were also higher in those subjects with detectable EA antibodies than in those without (F = 6.26; df = 1,28, p < 0.025), with a significant difference seen only in males when the sexes were considered separately (F = 6.61; df = 1,12;p < 0.025). There was no significant correlation between MADRS scores and VCA titers (p > 0.1). The relationship between MADRS scores and EA antibody titers observed in our newly admitted depressed patients may reflect EBV reactivation occurring in more severely depressed subjects. If further work establishes that this is the case, then this may offer a useful clinical model for exploring the im50-
40
o o
g
8 @
i.H tr
o 30 o o0 oo rr
$
o
oo
o 20
10
0
o
I <20
I
I 40
l
i 160
Reciprocal EA Titre
Figure 1: MADRS depression scores as a function of EA antibody tite~ (shown as reciprocal value) in ~,nale (o) and female (e) depressed patients. See text for results of Spearman correlation.
Correspondence
am~. PSYClIIATRY 199| ;29:611~-625
munologic consequences of depression. We have diseussed these issues at length elsewhere (Cooke et al 1989). In our sample, exposure to medication in the week prior to admission appeared not to affect EBV tilers, but our naturalistic and uncontrolled design did not allow us to rule out possible effects of age, medical condition, or earlier medication use in our drug-free subjects. To our knowledge, the effects of psychotropic medication on EBV antibody titers have not been reported. Future studies examining the relationship between EBV titers and depression will need to have carefully matched controls. Robert G. Cooke a'b Jerry J. Warsh a'b'c.e Gary M. Hase~)a'b Bernadette J. M. McLaughlin d'f Thecla Jornae "b
~Clarke Institute of Psychiatry 250 College Street Toronto, Ontario M5T IR8. Departments of bpsychiatry, ~Pharmacology and dMicrobiology, and qnstitute of Medical Science, University of Toronto. fLaboratory Services Branch, Ontario Ministry of Health, Toronto, Ontario.
References Albrecht J, Helderman JH, Schlesser MA, Rush AJ (1985): A controlled study of cellular inmmne function in affective disorders before and during somatic therapy. Psychiatry Res 15:185-193. Allen AD, Tilkian SM (1986): Depression correlated with cellular immunity in Systemic immunodeficient EpsteinBan"Virus Syndrome.,(SIDES). J Clin Psychiatry 47:133. American Psychiatric Association (1980): Diagnostic and Statistical Manual of Mental Disorders, Third Edition, (DSM HI). Washington, DC: APA, pp 210-218. Cooke RG, The psychiatrist and c.Vaonic fatigue syndrome. (1991): Can Dis Wkly Rep (supplement) (in press). Cooke RG, Langlet F, McLaughlin BJM (1988): Age-Specific prevalence of Epstein-Barr Virus antibodies in adult patients with affective&sorders.J Clin Psychiatry 49:361. Cooke RG, Warsh J J, Hasey GM ~,1989): Epstein-Barr Virus as a cause of autoimmtme disease and other medical morbidity in patients with affective disorders. Med Hypotheses 29:177.
623
Denney DR, Stephenson LA, Pvnick EC, W e l ~ RA ( 1 ~ ) : Lymphocyte subclasses and d e ~ . J Abnoem Psychol 94:499-502. Glaser R, Strain EC, Tart KL, Holliday/E, RL, Kiecolt-Glaser IlK (1985): Changes in Epstein-Barr virus antibody titers associated with aging (421~). Proc Soc Exp Bill Med 179:352. Irwin M, Gillin JC (1987): Impaired rialto'at killer cell activity among depressed patients. P~chiawy Res 20: t 8 i182. Keams NP, Cmickshank CA, McGuig-an KI, Riley SA, Shaw SP, Snaith RP (1982): A comparison of depression rating scales. Br J Psychiatry 141:45. King DJ, Cooper SI, Earle JAP, et at (1985): A survey of serum antibodies to eight common ~ s in psychiatric patients. Br J Psychiatry 147:137. Kronfol Z, Silva J Jr, Greden J, Dembinski S, Gardner R, Carroll B (1983): Impaired lymphocyte ~ d c a in depressive illness. Life Sci 33:241-247. Krordol Z, NasF,dlah HA, Chapman S, House ]D (1985): Depression, cortisol metabolism ~ d iymphocytopenia. d Affective Disord 9:16o- 173. Merino F, Henle W, Ramirez-Ihtque P (1986): Chronic active Epstein-Barr virus infection in patients with Chediak-Higashi syndrome. J Clin lmmunol 6:299-305. Montgomery SA, Asberg M (1979): A new depression scale designed to be sensitive to change. Br J Psychiatry 134:382. Pitts FN, Allen RE, H a r ~ ~/~, Alien AD (1987): Immunodeficiency t3 ia~,-pesviruses causes depression. Paper presented at me l,~'lh AnvxmIMeeting of the American Psychiatric Association, Chicago, May 9-14. Rinaldo CR Jr, Kingsley LA, Lyter DW, et al (1986): Association of ttTLV-III vA~ Epstein-Barr virus infection and abnormalities ofT lymphocytes in homosexual men. J Infect Dis 154:556. Salit ~ (1985): Sporadic postinfectious neuromyasthenia. Can Med Assoc J 133:659. Schle:~er SJ, Keller SE, Meyerson AT, Raskm MJ; Davis KL, Stein M (1984): Lymphocyte function in major depressive disorder. Arch Gen Psychiatry 41:4~ ~86. Schleifer SJ, Keller SE, Bond RN, Cohen J, Stein M (1989): Major depressive disorder and immunity. Role of age, sex, severity, and hospitaliTation. Arch Gen Psychiatry 46:81-87. Sumaya CV, Henle W, Henle G, Smith MHD, LeBlanc D (1975): Seroepidemiologic study of Epstein-Ban" infections in a rural community, J Infect Dis 131:403. Swanston W, MahonyJ, McLaughlin B, Chemesky M (1986): Assessment of serologic markers for Epstein-Barr virus. Diagn Microbiol Infect Dis 5:235. Syvalahti E, Eskola J, Ruuskanen O, Teijo L (1985): Nonsupp~ssion of cortisol in depression and immune function. Prog Neuropsychophannacol Biol Psychiatry 9:413422 Thorley-Lawson DA (1988): Basic virological aspects of Epstein-Barr virus infection. Semin Hematol 25:247260.