Immunological differences between patients with major depression and somatization syndrome

Psychiatry Research 105 Ž2001. 165᎐174

Immunological differences between patients with major depression and somatization syndrome Winfried Rief a,d,U , Florian Pilger a , Daniel Ihle a , Eugene Bosmans c , Belinda Egyedb, Michael Maes b a

b

Klinik Roseneck, Center for Beha¨ ioral Medicine, Prien am Chiemsee, Germany Department of Psychiatry and Neuropsychology, Uni¨ ersity Hospital of Maastricht, Maastricht, The Netherlands c Eurogenetics, Tessenderlo, Belgium d Uni¨ ersity of Marburg, Department of Clinical Psychology and Psychotherapy, Marburg, Germany Received 20 February 2001; received in revised form 12 September 2001; accepted 26 September 2001

Abstract There is some evidence that major depression is accompanied by activation of the inflammatory response system ŽIRS.. There is also evidence that proinflammatory cytokines and induction of IRS activation are associated with sickness behavior in experimental animals. However, no research has examined the IRS in somatization disorder. The aim of this study was to examine possible immunological differences between major depression, somatization and healthy controls. We measured the following IRS variables in patients with major depression Ž n s 36., somatization syndrome ŽSSI-8; n s 37., major depression and somatization Ž n s 40. and healthy controls Ž n s 37.: interleukin-6 ŽIL-6.; interleukin-1-receptor-antagonist ŽIL-1RA.; plasma soluble interleukin-6 receptor ŽIL-6R.; soluble suppressorrcytotoxic antigen ŽCD8.; leukemia inhibitory factor ŽLIF-R.; and Clara cell protein ŽCC16., an endogenous anticytokine. Serum CD8 concentrations were significantly increased in patients with major depression compared with concentrations in patients with somatization syndrome, whereas concentrations in normal controls were intermediate between those of the two groups of patients. Serum CC16 was significantly lower in major depression than in healthy controls. The highest CC16 scores were found in patients with somatization syndrome. Somatizing patients have significantly lower serum IL-6 values than normal controls and depressed patients. The present results indicate Ž1. an activation of the IRS in depression with signs of T-cell activation Žincreased CD8.,

Abbre¨ iations: IRSs inflammatory response system; IL s interleukin; SSIs Somatic Symptom Index; IL-1-RAs interleukin-1receptor-antagonist; IL-6R s plasma soluble interleukin-6 receptor; LIF-Rs leukemia inhibitory factor; CC16s Clara cell protein; TNFs tumor necrosis factor; SOMSs Screening for Somatoform Symptoms; BDI s Beck Depression Inventory; SCIDs Structured Clinical Interview for DSM-IV; IDCLs International Diagnostic Checklists U Corresponding author. Department of Clinical Psychology and Psychotherapy, University of Marburg, Gutenbergstr. 18, D-35032 Marburg, Germany. E-mail address: [email protected] ŽW. Rief.. 0165-1781r01r$ - see front matter 䊚 2001 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0 1 6 5 - 1 7 8 1 Ž 0 1 . 0 0 3 3 8 - 9

166

W. Rief et al. r Psychiatry Research 105 (2001) 165᎐174

monocytic activation ŽIL-1RA. and a lowered anti-inflammatory capacity of the serum Žlower CC16. and Ž2. an immune alteration in somatizing syndrome, such as monocytic activation Žincreased IL-1RA. and indicators of lowered T-lymphocytic activity Žlowered CD8 and IL-6.. These results suggest different immune alterations in somatization syndrome and depression. 䊚 2001 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Immunology; Depression; Somatization; Cytokines; Lymphocytes; Illness behavior

1. Introduction The inflammatory response system ŽIRS. and the concentration of cytokines are closely related to sickness behavior. The administration of peripheral cytokines, such as interleukin-1 ŽIL-1., in animals produces reductions of activity, exploration and social interaction, alterations in sleep, and other signs of illness ŽDantzer et al., 1998; Anisman and Merali, 1999.. Moreover, other results demonstrate that the peripheral administration of IL-1 leads to an increase in pain sensitivity ŽWatkins et al., 1995.. These sickness behaviors are prominent features of depression and, in a more chronic version, are key variables in severe somatization. There is evidence that major depression is associated with alterations of the inflammatory response system and of cytokines ŽBerk et al., 1997; Frommberger et al., 1997; Kronfol, 1999; Maes et al., 1995, 1997a; Sluzewska et al., 1996.. IRS activation in major depression may be indicated by increased numbers of leukocytes, monocytes, neutrophils, activated T-lymphocytes and increased numbers of neopterin and prostaglandins. Furthermore, an increased in vivo secretion of proinflammatory cytokines such as IL-6 and IL-8 was found in depressed patients. There is also evidence for an association between inflammatory processes and states of distress. Chronic distress in animals stimulates an increase of plasma IL-1, IL-2 and IL-6, a response that has been partly replicated in humans Žfor overview, see Black and Berman, 1999.. Experimental stress, which induces feelings of uncontrollability, can also influence IL-6 concentrations ŽPeters et al., 1999.. Further, IL-6 was found to be elevated in posttraumatic stress disorder ŽMaes et al., 1999.. The present study aims to replicate and extend

the findings on the IRS, cytokines and depression. We postulate an activation of the IRS in major depression. Moreover, we examine immunological variables in somatization. Somatization is a syndrome characterized by multiple physical symptoms not adequately explained by organic causes. These symptoms are very common and are frequent reasons for visits to the doctor ŽRief et al., 1996, 2001.. Patients with somatization tend to catastrophize physical symptoms, have a reduced capacity to tolerate physical discomfort and express a self-concept of being weak and not able to tolerate stress ŽRief et al., 1998a.. Physiological over-arousal can be present ŽRief and Auer, 2001; Rief et al., 1998b.. One central feature of this syndrome is illness behavior, which has an enormous impact on health care costs. Moreover, cytokines can stimulate inflammatory pain ŽSchaefer and Stein, 1999. and pain symptoms are the most frequent symptoms of somatization. However, psychobiological investigations of somatization are rare. In the following study, we examined signs of activation of the monocytic and T-lymphocytic arm of cell-mediated immunity, such as IL-6, soluble IL-6 receptor ŽIL-6R., plasma IL-1 receptor antagonist ŽIL-1RA., soluble suppressorrcytotoxic antigen ŽCD8., leukemia inhibitory factor receptor ŽLIF-R. and Clara cell protein ŽCC16.. Leukemia inhibitory factor ŽLIF. is a cytokine involved in the survival, differentiation and regeneration of sympathetic, sensory and motor neurons. The concentration of LIF-R has also been shown to be of importance in psychosomatic disorders Že.g. eating disorders; Monteleone et al., 1999. and has associations with pain perception ŽBanner et al., 1998.. Studies of CD8, CC16, and IL-1RA provide further information on Tlymphocytes and inflammatory response. IL-1RA

W. Rief et al. r Psychiatry Research 105 (2001) 165᎐174

167

is a naturally occurring molecule that is released in vivo during inflammation and is produced by activated cells of the monocytermacrophage lineage. It inhibits the biological activities of IL-1 at target cells, such as T-cells and fibroblasts. The CD8 molecule or suppressorrcytotoxic T-cell antigen is secreted by activated T-lymphocytes, such as activated CD8q T-cells. The Clara cell 16 kD protein ŽCC16. is a natural immunosuppressor and anti-inflammatory secretory protein. CC16 has been related to psychiatric illnesses Že.g. schizophrenia; DeJongh et al., 1998.. A shortcoming of many studies in psychobiology is the comparison of a clinical group with a group of healthy controls, which allows no interpretation of psychiatric specificity. Therefore, we compared multiple clinical groups and a control group.

symptoms ŽEscobar et al., 1989.. In an empirically derived approach to the definition of somatization, we found that a cut-off score of seven to eight symptoms is a useful way of differentiating patients with high and low levels of disability ŽRief and Hiller, 1999.. Ten of the 37 patients with somatization syndrome without major depression Ž27%. met full criteria for somatization disorder, while 15 of the 40 patients with somatization and comorbid major depression fulfilled all criteria for DSM-IV somatization disorder Ž37.5%., indicating that comorbidity may be associated with severity. Patients with major depression without somatization did not meet the criteria for weak somatization syndrome SSI-4r6 Žfour symptoms for men, six symptoms for women; Escobar et al., 1989.. Table 1 includes age, sex distribution and other characteristics of the groups.

2. Methods

2.2. Procedure

2.1. Subjects Participants of the clinical groups were inpatients of the Roseneck Center for Behavioral Medicine, a German inpatient treatment center for psychiatric and psychosomatic disorders. After a standardized diagnostic interview, patients were divided into three groups: 36 patients with major depression according to DSM-IV; 37 patients with somatization syndrome SSI-8; and 40 patients with major depression and somatization syndrome. The data for the patients were compared with those obtained in a healthy control group Ž n s 37., consisting mainly of employees of the hospital. Exclusion criteria were severe substance abuse or addiction, neurological disorders, immunological disorders and acute infections. Except for 13 women who were taking oral contraceptives, we excluded persons taking immunomodulating medication Že.g. cortisone therapy, psychopharmacotherapy.. For the definition of the somatization syndrome, the criteria for somatization disorder are too strict and do not represent the health-care relevance of patients with multiple somatoform

All consecutively admitted patients were prescreened with two self-rating scales: the Screen for Somatoform Symptoms ŽSOMS. and the Beck Depression Inventory ŽBDI. Žsee below.. Patients with more than eight unexplained physical symptoms or a BDI score of 20 or above were invited to an interview. A standardized psychiatric assessment took place Žsee below., in which participants were also told about the study and signed an informed consent form. In the following days, a physician ŽFP. visited the participant at 07:00 h and took blood samples of 2 = 7 cm3. Subjects were instructed to stay in bed until the blood sample had been taken. At 08:00 h, blood samples were centrifuged and stored at y70 ⬚C until thawed for assay. Between the interview session and the blood sampling, the participants filled out some questionnaires. All persons in the control group were screened for psychiatric disorders using the International Diagnostic Checklist. Three members of the healthy control group were excluded due to BDI depression scores above 7, indicating a substantial degree of depressive symptomatology. Controls

W. Rief et al. r Psychiatry Research 105 (2001) 165᎐174

168

were instructed to stay at home in bed until one of the investigators ŽFP. arrived at 07.00 h to take the blood sample. For this reason, the participants informed FP about their situation at home and gave him a key to their house on the day before the investigation.

2.3. Assessment instruments 2.3.1. Psychiatric inter¨ iew The Structured Clinical Interview for DSM-IV ŽSCID. ŽGerman version: Wittchen et al., 1997. was used to assess present and lifetime diagnoses.

Table 1 Sociodemographic and psychpathological variables

n

Somatization without depression

Somatization and depression

Depression with- Controls out somatization

37

40

36

Mean Age Gender Ž% female. Working disability Žweeks during last 12 months.a No. of DSM-IV somatization symptoms Žinterview. SOMS total score BDI No. of present state diagnoses Panic disorder

37

S.D.

Mean

S.D.

Mean

S.D.

43.46 65% 11.45

Ž12.12.

Ž10.30. Ž18.18.

42.25 64% 12.40

Ž10.33.

Ž16.69.

40.33 75% 19.00

9.68

Ž2.39.

11.65

Ž3.26.

19.08 14.24 2.08

Ž6.08. Ž5.41. Ž0.83.

22.03 29.25 3.60

Ž6.44. Ž7.40. Ž1.46.

Mean

Ž11.8.

Ž15.90.

Ž1.4.

F s 1.8, ns Chi2 s 5.2, ns F s 4.2UU

3.53

Ž1.80.

0.9

Ž1.0.

F s 181UUU

12.86 26.78 2.39

Ž6.66. Ž8.20. Ž1.13.

2.4 2.7 0.02

Ž2.4. Ž3.1. Ž0.2.

F s 86UUU F s 138UUU F s 78UUU

49%

33%

5%

15%

22%

0%

OCD

11%

15%

6%

0%

PTSD

5%

20%

8%

0%

Body mass index Illness behavior Žexpert rating; 1 . . . 4. SCL-90-R scores: Somatization Obsessive-compulsive Interpersonal sensitivity Depression Anxiety Anger-hostility Phobic anxiety Paranoid ideation Psychoticism General symptomatic index Positive symptoms Stress index of complaints a

0%

25.35 2.21

Ž4.67. Ž0.39.

26.74 2.16

Ž6.36. Ž0.45.

24.85 2.14

Ž4.47. Ž0.54.

25.0 2.72

1.31 1.21 1.02 1.21 1.17 0.66 0.68 0.99 0.55 1.03

Ž0.67. Ž0.65. Ž0.61. Ž0.53. Ž0.62. Ž0.53. Ž0.74. Ž0.69. Ž0.40. Ž0.44.

1.76 1.95 1.87 2.19 1.87 1.05 1.25 1.57 1.08 1.70

Ž0.67. Ž0.68. Ž0.80. Ž0.63. Ž0.84. Ž0.66. Ž0.94. Ž0.97. Ž0.60. Ž0.50.

1.16 1.78 1.80 2.06 1.43 1.24 1.11 1.51 0.93 1.49

Ž0.67. Ž0.77. Ž0.97. Ž0.77. Ž0.71. Ž0.87. Ž1.03. Ž0.94. Ž0.58. Ž0.60.

0.28 0.24 0.24 0.23 0.18 0.18 0.08 0.25 0.05 0.20

50.81 1.78

Ž15.92. Ž0.35.

65.22 2.31

Ž10.83. Ž0.43.

59.00 2.20

Ž14.31. Ž0.56.

16.17 1.08

Only employed persons. UU P- 0.01; UUU P- 0.001.

S.D.

37.8 50% 1.0

32%

Social phobia

Significance

Chi2 Žonly clinical groups. s 12, ns Chi2 Žonly clinical groups. s 9, ns Chi2 Žonly clinical groups. s 4, ns Chi2 Žonly clinical groups. s 6, ns Ž4.3. F s 1.0, ns Ž0.22. F s 16UUU

Ž0.28. Ž0.27. Ž0.24. Ž0.22. Ž0.20. Ž0.23. Ž0.12. Ž0.28. Ž0.06. Ž0.17.

F s 27.8UUU F s 39.1UUU F s 30.6UUU F s 64.6UUU F s 31.5UUU F s 14.9UUU F s 11.4UUU F s 15.8UUU F s 25.2UUU F s 54UUU

Ž12.38. F s 69.5UUU Ž0.11. F s 49.2UUU

W. Rief et al. r Psychiatry Research 105 (2001) 165᎐174

As both interviewers ŽFP, DI. were trained and experienced in psychiatric diagnoses, the diagnostic procedure was facilitated by using the International Diagnostic Checklist ŽIDCL; Hiller et al., 1997. as an interview guideline after the interviewer had conducted at least 10 structured interviews using the SCID. 2.3.2. Somatization index The criteria for DSM-IV somatization disorder include a list of 33 medically unexplained physical symptoms. The DSM-IV somatization index is the total score of symptoms out of these 33 bodily complaints. 2.3.3. Illness beha¨ ior To assess aspects of illness behavior, the patient’s physician was asked to rate variables of illness behavior after studying the patient’s record and after hisrher physical examination. The physician rated seven items wexamples: the patient is seeking medical help Ž1. more than necessary; Ž2. appropriately and Ž3. less than necessary x. These items were summed to provide a global score for illness behavior. Thus, medium scores indicate appropriate illness behavior, low scores indicate excessive illness behavior, and high scores indicate avoidance of seeking help. 2.3.4. Screen for Somatoform Symptoms (SOMS) The SOMS ŽRief et al., 1997. is a self-rating scale that inquires about approximately 53 physical symptoms during the last 2 years. The symptoms include all somatoform symptoms listed in the classification of DSM-IV somatization disorder. Additional symptoms from the ICD-10 criteria for somatization disorder and somatoform autonomic dysfunction are also included. For the present purpose, we use the somatization index according to DSM-IV. This index includes all endorsed symptoms from the list of 33 somatoform symptoms mentioned in DSM-IV somatization disorder. The retest reliability of such a somatization index is rt t s 0.85 Ž72 h. and the correlation of the somatization index according to a self-rating scale with a diagnostic interview is r s 0.75 ŽRief et al., 1997..

169

2.3.5. Depression and other psychopathological ¨ ariables

The psychometric measure for depression was the Beck Depression Inventory ŽBDI; Beck et al., 1961.. Moreover, the 90-item version of the Symptom Checklist-90 Revised ŽSCL-90-R; Derogatis, 1994; Rief and Fichter, 1992. revealed different aspects of psychopathology with nine specific dimensions and three general scores. 2.3.6. Analysis of blood samples The blood samples were analyzed for the following variables: interleukin-1 receptor antagonist ŽIL-1RA., interleukin-6 ŽIL-6., interleukin-6 receptor ŽIL-6R., Clara cell protein ŽCC16., leukemia inhibitory factor receptor ŽLIF-R. and markers for cytotoxic T-lymphocytes ŽCD8.. All assays of the parameters were carried out at the same time and in the same run at Eurogenetics ŽTessenderlo, Belgium.. The immunological variables were measured by quantitative enzyme-linked immunosorbent assay ŽELISA. techniques, based on appropriate and validated sets of monoclonal antibodies. For all variables, the intra-assay coefficient of variation was lower than 8%. 2.3.7. Statistical analysis The psychometric variables, expert ratings for illness behavior and age were analyzed parametrically using one-way analysis of variance ŽANOVA.. As the immunological parameters did not show a normal distribution and outliers occurred, these data were analyzed using non-parametric statistics. In comparison to the frequently used log-transformation, the non-parametric approach was preferred as it is based on more conservative assumptions1. Extreme outliers Ž) 3 S.D.. were considered as missing values. For all immunological parameters, the rate of missing values was below 10% . If the overall Kruskal᎐Wallis test revealed significant group differences for the immunological parameters, we made subsequent pairwise comparisons of all combinations. After analyses of group differ-

1

Results of the log-transformed data can be requested from the first author.

W. Rief et al. r Psychiatry Research 105 (2001) 165᎐174

170

ences, correlational analyses were done using Spearman rank coefficients for exploratory reasons.

3. Results 3.1. Group differences As expected, the psychometric variables confirmed the differences of the four groups regarding depression and somatization Žsee Table 1.. Long periods of working disability days were associated with depression and somatization, being most pronounced in patients with both disorders. Both depression and somatization have substantial comorbidity rates for anxiety disorders, a fact that was reported frequently. The SCL-90-R confirmed the most severe pathology for patients with both disorders. Table 2 presents mean values for immunological parameters in the four groups and results of the Kruskal᎐Wallis tests. After testing whether the immunological variables of the 13 women taking contraceptives differed from those of 60 comparable women by means of Kruskal᎐Wallis tests, we did not exclude these persons, as none of the biological parameters differed significantly between women taking contraceptives and those who did not Žchi-square s 0.3᎐2.1, N.S... The major depressive patients showed the highest CD8 concentrations. Both groups of somatiz-

ing patients Ži.e. with and without comorbid depression. had significantly lower CD8 counts than the major depressive patients. Healthy controls had scores intermediate between those of the pure depression and the somatization groups. The results for interleukins revealed some differences, though less pronounced than those for CD8. IL-6 was reduced in patients with somatization syndrome, who differed significantly from all other groups. The receptor variable IL-6R, contrary to IL-6, had lower concentrations in the depression group than in the somatization group. The receptor antagonist variable IL-1RA was significantly elevated in all patient groups compared with controls. The most significant group differences were found for the anti-cytokine Clara cell protein CC16 Ž ␹ 2 s 43.1.. Patients with somatization syndrome Žwith or without depression. had significantly higher CC16 concentrations than controls, while major depressive patients Žwithout somatization. showed a trend Ž Ps 0.10. toward lower concentrations than those in control subjects. To allow a comparison with studies using the classical definition of somatization disorder according to DSM-IV, we reanalyzed our data comparing subjects with full somatization disorder Žfull criteria according to DSM-IV, comorbidity with depression was accepted, n s 25. and subjects with major depression Žaccording to DSMIV, without somatization disorder, n s 61.. Both clinical groups still differed in terms of CD8

Table 2 Immunological variables in clinical groups

IL-1RA IL-6 IL-6R CC16 LIF-R CD-8

Somatization A

Somatization and depression B

Major depression C

Controls D

Mean

S.D.

Mean

S.D.

Mean

S.D.

Mean

S.D.

0.244 3.36 114.74 37.65 1.68 265.21

0.502 4.92 30.34 20.05 0.77 101.0

0.200 3.69 104.78 39.08 1.34 306.75

0.197 3.84 46.09 15.92 0.69 105.92

0.144 4.51 96.12 19.62 1.35 402.29

0.097 5.63 29.66 10.25 0.69 154.9

0.105 3.91 95.5 24.91 1.42 348.93

0.129 3.27 27.9 12.71 0.68 112.57

Remarks: significance of Kruskal᎐Wallis ŽK᎐W. test: brackets: P - 0.10.

U

K-W chi 2

16.7UU AD, BD, CD 9.7U AB, AC, AD 8.9U AB, AC, ŽAD. 43.1UUU AC, AD, BC, BD, ŽCD. 4.9 19.9UUU AB, AC, AD, BC

P- 0.05, UU P- 0.01, UUU P- 0.001; pairwise comparisons: P- 0.05; in

W. Rief et al. r Psychiatry Research 105 (2001) 165᎐174

counts Ž ␹ 2 s 7.6, P- 0.01., CC16 Ž ␹ 2 s 8.4, P0.01. and IL-6R Ž ␹ 2 s 8.4, P- 0.01., confirming the results reported above. 3.2. Associations of immunological parameters with psychological ¨ ariables In exploratory analyses, Spearman rank correlations of immunological with psychological variables were computed. Only when depression and somatization were characterized by immunological alternations in the same direction Žsuch as for IL-1RA. did a significant association between BDI depression scores and indicators of monocytic activation emerge. However, IL-1RA showed positive correlations with all variables of psychopathology, thus reflecting a non-specific association. No substantial correlations were found between illness behavior and immunological parameters; therefore, these results were not included in Table 3. The number of somatoform symptoms was negatively associated with IL-6 and CD-8 values, and positively associated with CC16, IL-1RA and IL-6R values.

4. Discussion Some studies have addressed immunological aspects of depression and have demonstrated an activation of parameters of the inflammatory response system Že.g. Berk et al., 1997; Sluzewska et al., 1996; Maes et al., 1997a.. Increased numbers of activated T-lymphocytes, leukocytes, monocytes, and neutrophils, but also increased in vivo secretion of proinflammatory cytokines Žsuch as IL-6., have been reported in depression. For somatization, however, a condition frequently co-

171

morbid with depression and one of the most frequent reasons for doctor visits, investigations on psychoneuroimmunological aspects are extremely rare. Illness behavior, one of the major features of somatization as well as of depression, was investigated in some immunological studies. Crestani et al. Ž1991. observed social withdrawal, general sickness behavior and reduction of social exploratory behavior in laboratory animals after the application of interleukin-1. Further studies confirmed an association of illness behavior with cytokine concentrations ŽBluthe et al., 1995, 1999.. These results suggest that further investigations of immunological aspects of somatization would be of interest. In our study, we analyzed blood samples and psychological variables of 150 persons from three clinical groups and a control group. Our major finding is that the immunological states of somatizing patients are significantly different from those of depressive patients, despite the fact of high comorbidity rates between the two disorders. This difference was found not only for patients with somatization syndrome, but also for patients who met the full diagnostic criteria for somatization disorder. Lower CD8 and IL-6 levels may be indicators of reduced lymphocytic activity in somatization. A higher CC16 concentration in somatization is a further indicator of reduced inflammatory capacity in this group, but the functional meaning of these findings has to be further tested. Our patients with major depression had elevated CD8 counts and reduced CC16 concentrations in the serum. This confirms in part earlier findings of indices of an activated IRS in depressive patients

Table 3 Correlations of immunological parameters with psychological variables ŽSpearman Rank. IL-1RA Depression ŽBDI. No. of somatoform symptoms Anxiety ŽSCL-90-R. Phobic anxiety ŽSCL-90-R. U

UUU

0.30 0.22UU 0.23UU 0.25UU

P- 0.05; UU P- 0.01; UUU P- 0.001.

IL-6

IL-6R

CC16

LIF-R

CD8

0.05 y0.22UU y0.02 y0.13

0.02 0.18U 0.02 0.02

0.07 0.37UU 0.11 0.09

y0.13 y0.01 y0.10 y0.06

0.13 y0.20U 0.04 0.09

172

W. Rief et al. r Psychiatry Research 105 (2001) 165᎐174

Že.g. Maes et al., 1997a.. A further difference between somatizing patients and depressive patients Žwithout somatization. was found for IL6R, with higher concentrations being found for somatizing patients. Situational and psychological factors also influence immunological parameters. It has been demonstrated that stressors Žsuch as academic stress . can lead to an increase of CD3, CD4 and CD8 T-cell subsets ŽGriffiths et al., 1997; Lin et al., 1998; Song et al., 1999.. Recently, Peters et al. Ž1999. investigated the effects of mental effort and uncontrollability on immunological parameters such as CD8q , IL-4 and IL-6. They found that mental effort had a stimulating effect on immunological parameters such as CD8q count and IL-6 concentration. Petrie et al. Ž1998. demonstrated that the expression of emotions can lead to an increase of CD4 T lymphocyte levels. Cohen et al. Ž1999. found that higher psychological stress before a viral challenge was associated with higher IL-6 lavage concentrations. While these psychological effects may be associated with an activation of the IRS, antidepressants, on the other hand, can have immunosuppressive effects, and may support a partial immunological normalization ŽMaes et al., 1997a; Miller and Lackner, 1989.. It is unclear, however, how immunological alterations interact with psychopathological features of somatization. Some studies reveal that physical exercise leads to an increase of proinflammatory cytokines ŽDufaux and Order, 1989.. As patients with somatization tend to avoid physical exercise, physical deconditioning may be associated with a down-regulation of proinflammatory processes. Moreover, as the immunological reactions to distress may vary depending on the nature of the challenging events and the interaction with personality characteristics ŽEvans et al., 2000., stress can have different consequences in patients with somatization and patients with depression. However, such a theory is quite speculative. Our major immunological results in regard to depression are somewhat comparable to those found for schizophrenic patients, especially for the variables CD8 and CC16 ŽLin et al., 1998;

Maes et al., 1997b.. In contrast to earlier findings, however, we could not confirm increased IL-6 concentrations in depressive patients compared with control subjects. As expected, major depressive patients had the highest mean scores for IL-6, but the difference from the control subjects did not reach statistical significance. The consideration of anti-inflammatory agents such as CC16 has a short history in psychoimmunology. Therefore, our results on CC16 are promising. DeJongh et al. Ž1998. suggest that most of the effects of CC16 are achieved by modulating the cytokine network and its effectors. Obviously, CC16 can also be a prominent and sensitive variable differentiating psychiatric groups, and it has enormous discriminating power between somatization and depression. It seems to be promising to investigate the interactions of CC16 with further psychological features in longitudinal studies. Moreover, we found that IL-1RA was elevated in all patient groups compared with the control group. Correlational analyses revealed that IL1RA was associated with various psychological conditions such as depression, somatization, and anxiety. Maier and Watkins Ž1998. demonstrated in animal studies that the external application of IL-1RA before inescapable shock blocked the enhancement of fear conditioning that is normally produced by this treatment. These conditioning processes may play a role in the development of various psychological disorders. In summary, the strength of this study was its consideration of various clinical groups to allow the analysis of psychiatric specificity. However, some limitations reduce the generalizability of the results. First, further studies are needed to replicate our results. Second, we used current state diagnoses, and the consideration of lifetime diagnoses could highlight further aspects of the reported results. Moreover, there is a major need to investigate the pathways between immunological processes and psychopathological features of psychiatric disorders. Finally, the functional meaning of the differences in the concentration of immunological parameters has to be addressed. The concentration of parameters in peripheral blood is not an absolute indicator of

W. Rief et al. r Psychiatry Research 105 (2001) 165᎐174

immunological activity but depends on trafficking characteristics, shifts between cellular and humoral immunity, and other factors. These questions should be answered in subsequent studies. References Anisman, H., Merali, Z., 1999. Anhedonic and anxiogenic effects of cytokine exposure. Advances of Experimental Medicine and Biology 461, 199᎐233. Banner, L.R., Patterson, P.H., Allchorne, A., Poole, S., Woolf, C.J., 1998. Leukemia inhibitory factor is an anti-inflammatory and analgesic cytokine. Journal of Neuroscience 18, 5456᎐5462. Beck, A.T., Ward, A.H., Mendelson, M., Mock, F., Erbaugh, F., 1961. An inventory for measuring depression. Archives of General Psychiatry 4, 561᎐571. Berk, M., Wadee, A.A., Kuschke, R.H., O’Neill-Kerr, A., 1997. Acute phase proteins in major depression. Journal of Psychosomatic Research 43, 529᎐534. Black, P.H., Berman, A.S., 1999. Stress and inflammation. In: Plotnikoff, N.P., Faith, R.E., Murgo, A.J., Good, R.A. ŽEds.., Cytokines, Stress and Immunity. CRC Press, Boca Raton, FL. Bluthe, R.M., Beaudu, C., Kelle, K.W., Dantzer, R., 1995. Differential effects of IL-1ra on sickness behavior and weight loss induced by IL-1 in rats. Brain Research 677, 171᎐176. Bluthe, R.M., Castanon, N., Pousset, F., Bristow, A., Ball, C., Lestage, J., Michaud, B., Kelley, K.W., Dantzer, R., 1999. Central injection of IL-10 antagonizes the behavioural effects of lipopolysaccharide in rats. Psychoneuroendocrinology 24, 301᎐311. Cohen, S., Doyle, W.J., Skoner, D.P., 1999. Psychological stress, cytokine production, and severity of upper respiratory illness. Psychosomatic Medicine 61, 175᎐180. Crestani, F., Segoy, F., Dantzer, R., 1991. Behavioral effects of peripherally injected interleukin-1: role of prostaglandin. Brain Research 542, 330᎐335. Dantzer, R., Bluthe, R.M., Laye, S., Bret-Dibat, J.L., Parnet, P., Kelley, K.W., 1998. Cytokines and sickness behavior. Annals of the New York Academy of Sciences 840, 586᎐590. DeJongh, R., Vranken, J., Kenis, G., Bosmans, E., Maes, M., Stans, G., DeLey, M., Heylen, R., 1998. Clara cell protein: concentrations in cerebrospinal fluid, serum and amniotic fluid. Cytokines 10, 441᎐444. Derogatis, L.R., 1994. SCL-90-R. Administration, Scoring and Procedures Manual. NCS, Minneapolis. Dufaux, B., Order, U., 1989. Plasma elastase-alpha 1-antitrypsin, neopterin, tumor necrosis factor and soluble interleukin-2 receptor after prolonged exercise. International Journal of Sports Medicine 10, 434᎐438. Escobar, J.I., Rubio-Stipec, M., Canino, G., Karno, M., 1989. Somatic symptoms index ŽSSI.: a new and abridged somati-

173

zation construct: prevalence and epidemiological correlates in two large community samples. Journal of Nervous and Mental Disease 177, 140᎐146. Evans, P., Hucklebridge, F., Clow, A., 2000. Mind, Immunity and Health: The Science of Psychoneuroimmunology. Free Association Books, London᎐New York. Frommberger, U.H., Bauer, J., Haselbauer, P., Fraulin, A., Riemann, D., Berger, M., 1997. Interleukin-6- ŽIL-6. plasma levels in depression and schizophrenia: comparison between the acute state and after remission. European Archives of Psychiatry and Clinical Neuroscience 247, 228᎐233. Griffiths, J., Ravindran, A.V., Merali, Z., Anisman, H., 1997. Neuroendocrine measures and lymphocyte subsets in depressive illness: influence of a clinical interview concerning life experiences. Psychoneuroendocrinology 22, 225᎐236. Hiller, W., Zaudig, M., Mombour, W., 1997. DSM-IV Checklisten. Hans Huber, Bern ŽCH.. Kronfol, Z., 1999. Depression and immunity: the role of cytokines. In: Plotnikoff, N.P., Faith, R.E., Murgo, A.J., Good, R.A. ŽEds.., Cytokines, Stress and Immunity. CRC Press, Boca Raton, FL. Lin, A., Kenis, G., Bignotti, S., Tura, G.J.B., DeJong, R., Bosmans, E., Pioli, R., Altamura, C., Scharpe, ´ S., Maes, M., 1998. The inflammatory response system in treatment-resistant schizophrenia: increased serum interleukin-6. Schizophrenia Research 32, 9᎐15. Maes, M., Bosmans, E., DeJongh, R., Kenis, G., Vandoolaeghe, E., Neels, H., 1997a. Increased serum IL-6 and IL-1 receptor antagonist concentrations in major depression and treatment resistant depression. Cytokines 9, 853᎐858. Maes, M., Bosmans, E., Kenis, G., DeJong, R., Smith, R.S., Meltzer, H.Y., 1997b. In vivo immunomodulatory effects of clozapine in schizophrenia. Schizophrenia Research 26, 221᎐225. Maes, M., Lin, A., Delmeire, L., VanGastel, A., Kenis, G., DeJongh, R., Bosmans, E., 1999. Elevated serum interleukin-6 ŽIL-6. and IL-6 receptor concentration in posttraumatic stress disorder following accidental man-made traumatic events. Biological Psychiatry 45, 833᎐839. Maes, M., Smith, R., Scharpe, ´ S., 1995. The monocyte-T lymphocyte hypothesis of major depression. Psychoneuroendocrinology 20, 111᎐116. Maier, S.F., Watkins, L.R., 1998. Cytokines for psychologists: implications for bidirectional immune-to-brain communication for understanding behavior, mood, and cognition. Psychological Review 105, 83᎐107. Miller, A.H., Lackner, C., 1989. Tricyclic antidepressants and immunity. In: Miller, H. ŽEd.., Depressive Disorders and Immunity. APA, Washington, DC, pp. 85᎐104. Monteleone, P., Maes, M., Fabrozzo, M., Tortorella, A., Lin, A., Bosmans, E., Kenis, G., Maj, M., 1999. Immunoendocrine findings in patients with eating disorders. Neuropsychobiology 40, 115᎐120. Peters, M.L., Godeart, G.L.R., Ballieux, R.E., Brosschot, J.F.,

174

W. Rief et al. r Psychiatry Research 105 (2001) 165᎐174

Sweep, F.C.G.J., Swinkels, L.M.J.W., van Vliet, M., Heijnen, C.J., 1999. Immune responses to experimental stress: effects on mental effort and uncontrollability. Psychosomatic Medicine 61, 513᎐524. Rief, W., Auer, C., 2001. Is somatization a habituation disorder? Physiological reactivity in somatization syndrome. Psychiatry Research 101, 63᎐74. Rief, W., Fichter, M.M., 1992. The symptom check list SCL90R and its ability to discriminate between dysthymia, anxiety disorders, and anorexia nervosa. Psychopathology 25, 128᎐138. Rief, W., Hessel, A., Braehler, E., 2001. Somatization symptoms and hypochondriacal features in the general population. Psychosomatic Medicine 63, 595᎐602. Rief, W., Heuser, J., Mayrhuber, E., Stelzer, I., Hiller, W., Fichter, M.M., 1996. The classification of multiple somatoform symptoms. Journal of Nervous and Mental Disease 184, 680᎐687. Rief, W., Hiller, W., 1999. Toward empirically based criteria for somatoform disorders. Journal of Psychosomatic Research 46, 507᎐518. Rief, W., Hiller, W., Heuser, J., 1997. SOMS ᎏ Das Screening fur Manual zum Fragebogen ¨ Somatoforme Storungen ¨ wSOMS-Screening for somatoform symptoms. Manual for the self-rating scalex. Hans Huber, Bern. Rief, W., Hiller, W., Margraf, J., 1998a. Cognitive aspects in hypochondriasis and the somatization syndrome. Journal of Abnormal Psychology 107, 587᎐595. Rief, W., Shaw, R., Fichter, M.M., 1998b. Elevated levels of

psychophysiological arousal and cortisol in patients with somatization syndrome. Psychosomatic Medicine 60, 198᎐203. Schaefer, M., Stein, C., 1999. Control of pain in peripheral tissues by cytokines and neuropeptides. In: Plotnikoff, N.P., Faith, R.E., Murgo, A.J., Good, R.A. ŽEds.., Cytokines, Stress and Immunity. CRC Press, Boca Raton, FL, pp. 261᎐270. Sluzewska, A., Rybakowski, J., Bosmans, E., Sobieska, M., Berghmans, R., Maes, M., Wiktorowics, K., 1996. Indicators of immune activation in major depression. Psychiatry Research 64, 161᎐167. Song, C., Kenis, G., Gastel, A., Bosmans, E., Lin, A., de Jong, R., Neels, H., Scharpe, ´ S., Janca, A., Yasukawa, K., Maes, M., 1999. Influence of psychological stress on immune-inflammatory variables in normal humans. Part II. Psychiatry Research 85, 293᎐303. Watkins, L.R., Maier, S.F., Goehler, L.E., 1995. Immune activation: the role of proinflammatory cytokines in inflammation, illness response, and pathological pain states. Pain 63, 289᎐302. Petrie, K.J., Booth, R.J., Pennebaker, J.W., 1998. The immunological effects of thought suppression. Journal of Personality and Social Psychology 75, 1264᎐1272. Wittchen, H.U., Schramm, E., Zaudig, M., Unland, H., 1997. SKID Stukturiertes klinisches Interview fur ¨ DSM-IV, Achse I, deutsche Version. wSCID-Structured Clinical Interview for DSM-IV, axis I, German Versionx. Hogrefe, Gottingen. ¨