Secondary Schizophrenia (112 page)

prevalence of schizophrenia and other spectrum dis-Increased IL-8 has been associated with some autoim-orders was 20.4%, which suggested a 15 to 20-fold
mune disorders including psoriasis
[46].

increased risk of these disorders given population-For each of the infections and immune abnormal-based estimates for this broader diagnostic outcome
ities discussed above, we must also consider the possi-

282

[30,
31]
. The effect appeared strongest for the first
bility that nonspecific pathogenic mechanisms might
Chapter 22 – Infection and schizophrenia

account for the findings. These include hyperther-

Suggestions for future research

mia, a known cause of neural tube defects in animal
First, we believe it is essential to replicate these find-studies
[47],
cold, flu, or analgesic remedies, which
ings in independent samples and to identify additional
may have teratogenic effects
[47],
and stress, which
infections that may play etiologic roles in schizophre-induces hypothalamo-pituitary-adrenal axis activa-nia. In order to accomplish this, we believe it is essen-tion that causes maldevelopment of the hippocampus
tial to employ larger samples utilizing newer technolo-and other brain structures and functions implicated in
gies to identify a wider array of infectious pathogens
the pathogenesis of schizophrenia
[48].

than currently exist, for example, high throughput
assays that can delineate the microbial genome. Several
large cohort studies are under way that will permit the
Studies of infection and schizophrenia

application of these approaches. Second, it is unlikely
in adult patients

that prenatal environmental exposures are sufficient in
and of themselves to result in schizophrenia in most
Although not the focus of this chapter, it is worth not-cases. Thus, we believe that future work is necessary
ing that a considerable number of studies have demon-to examine whether these infectious exposures inter-strated increased seroprevalence of toxoplasmosis in
act with susceptibility genes to increase vulnerability to
patients with schizophrenia
[49].
In a recent meta-this disorder. Intriguingly, several susceptibility genes
analysis, the combined odds ratio was 2.73 (95% CI
=

for schizophrenia that have been replicated in several
2.1–3.6). The finding was similar for both first-episode
studies appear to play important roles in fetal brain
patients and chronically ill patients. Although intrigu-development
[50, 51]
. Given the relatively small effects
ing, the studies reviewed are characterized by substan-of individual genes in previous association studies,
tial variability in methodological design, and rigor-the modest effect sizes observed in the positive stud-ous epidemiologic methods were generally not used
ies of prenatal infection, and the large samples nec-with regard to case and control ascertainment. Of
essary for studies of gene–environment interaction to
note, among the 23 studies reviewed, controls were
achieve adequate statistical power, we anticipate that
variously defined as “general population,” “hospital
the numbers of cases for these studies will need to be in
employees,” and “normal persons.” The use of indi-the hundreds or perhaps thousands in order to reveal
viduals who are screened out for all psychiatric disor-meaningful interactions; thus, the same large cohorts
ders and hospital employees is prone to the selection
alluded to above should be highly suitable for this
of a “super-normal,” control group, characterized by
purpose.

better lifestyle habits than a control group that represented the source population from which the cases
Translational approaches

were derived. Moreover, given that all of these studies are cross-sectional, the direction of cause and effect
Translational research approaches hold the promise
cannot be definitively determined. Lifestyle differences
of identifying pathogenic mechanisms by which these
secondary to the illness that could predispose to toxo-exposures increase schizophrenia risk. Animal mod-plasma infection may even have occurred among first-els have provided evidence that offspring of mice
episode patients, who may have been ill for some time
that were infected with influenza at day 9 of ges-before the diagnosis of schizophrenia.

tation exhibit deficits in prepulse inhibition (PPI)
in the acoustic startle response, alterations in open-field, novel object, and social interaction tests, and
improvement in PPI disruptions following adminis-

Implications for clinical practice

tration of psychotomimetic and antipsychotic medica-Research on prenatal etiologies of schizophrenia may
tions
[52].
PPI and other deficits were also observed
have important implications for prevention of this dis-when poly I:C, a potent immune activator, was admin-order, through the institution of public health mea-istered at this time in fetal development in the absence
sures. This work has the potential of identifying at-of I infection, suggesting that these aberrations were
risk individuals who experienced in utero insults, or
secondary to the maternal immune response. In
their sequelae, and who possess genetic vulnerability
another study, prenatal influenza appeared to dis-

283

to schizophrenia.

rupt neuronal migration, as evidenced by decreased
Organic Syndromes of Schizophrenia – Section 3

reelin-positive Cajal-Retzius cells in the cortex and
of these exposures can be prevented by relatively inex-hippocampus
[53].
Translational approaches may also
pensive public health measures; for example, influenza
facilitate the identification of new molecular targets
is preventable by vaccination of women of childbear-for psychopharmacologic intervention. In addition
ing age; the incidence of T. gondii can be markedly
to the correction of prenatal immune-mediated neu-reduced by standard hygienic practices such as avoid-rophysiologic and behavioral abnormalities, activa-ance of direct contact with cat feces and ingestion
tion of the maternal immune response was associated
of undercooked meat
[55];
several maternal/genital
with the postpubertal emergence of increased sensitiv-reproductive infections can be minimized by con-ity to locomotor stimulation following amphetamine
dom use and early identification and treatment; and
administration and increased amphetamine-induced
homocysteine levels can be normalized by ingestion of
striatal dopamine release
[54].

folate from the diet or vitamin supplements. Although
rubella has been virtually eliminated in the industri-

Summary and conclusions

alized world, it remains a significant problem in the
developing world, where the vaccine is not routinely
Despite a considerable amount of research on the
administered. Some of our greatest successes in global
underpinnings of schizophrenia, its etiology and
medicine in the twentieth century have been achieved
pathogenesis remain unknown. In utero infections
by the elimination of infectious diseases through vacci-are emerging as potentially important risk factors for
nation programs and improved sanitation. We believe
schizophrenia, in which neurodevelopmental influ-that similar public health initiatives have the poten-ences likely play an important role. Our group and oth-tial to reap similar rewards with regard to a reduc-ers have embarked on investigations aimed at identi-tion in the incidence of schizophrenia and potentially
fying these infectious risk factors and examining the
reveal new molecular targets for pharmacotherapeutic
mechanisms by which they increase vulnerability to
intervention.

this disorder.

This work has the potential to lead to strategies
aimed at preventing this disorder. Most of the pre-

Acknowledgments

natal infections identified to date are relatively com-This manuscript was supported by the following
mon during pregnancy, and these factors confer at
grants: NIMH 1K02MH65422–01 (A.S.B.), NIMH

least a twofold, and as high as a fivefold, increase in
1R01MH 63264–01A1 (A.S.B.), an NARSAD Inde-

risk of schizophrenia. Hence, the combined popula-pendent Investigator Award (A.S.B.), NICHD N01-tion attributable risk (PAR) (the proportion of cases in
HD-1–3334 (B. Cohn), and NICHD NO1-HD-6–3258

the population that could be prevented if these expo-

(B. Cohn). We wish to thank Catherine Schaefer,
sures were eliminated) is potentially substantial. For
PhD, Barbara Cohn, PhD, Barbara van den Berg, MD,
influenza alone, the PAR for early to mid-gestational
Michaeline Bresnahan, PhD, and Justin Penner, MA,
exposure may be as high as 10% to 15%
[13].
Many
for their contributions to this work.

284

Chapter 22 – Infection and schizophrenia

References

11. Van Den Berg B. J. (1984). The

Am J Psychiatry, 2005.
162
(4):
California child health and

767–73.

1. Bradbury T. N., Miller G. A.

development studies. In

Season of birth in schizophrenia:

19. Mortensen P. B.,

Handbook of Longitudinal

a review of evidence,

Nergaard-Pedersen B., Waltoft

Research, Mednick S. A., Harway

methodology, and etiology.

B. L.,
et al.
Toxoplasma gondii as a
M., and Finello K. (Eds.). New

Psychol Bull, 1985.
98
(3):569–94.

risk factor for early-onset

York: Praeger, pp. 166–79.

schizophrenia: analysis of filter

2. Lewis G., David A, Andréasson S.,

12. Susser E. S., Schaefer C. A., Brown

paper blood samples obtained at

Allebeck P. Schizophrenia and city

A. S., Begg M. D., Wyatt R. J. The

birth. Biol Psychiatry, 2007.
61
(5):
life. Lancet, 1992.
340
(8812):137–
design of the prenatal

688–93.

40.

determinants of schizophrenia

20. Frenkel J. K., Nelson B. M.,

3. Mortensen P. B., Pedersen C. B.,

study. Schizophr Bull, 2000.

Arias-Stella J. Immuno-

Westergaard T.,
et al.
Effects of

26
(2):257–73.

suppression and toxoplasmic

family history and place and

13. Brown A. S., Begg M. D.,

encephalitis – clinical and

season of birth on the risk of

Gravenstein S.,
et al.
Serologic

experimental aspects. Hum

schizophrenia. N Engl J Med,

evidence for prenatal influenza in

Pathol, 1975.
6
(1):97–111.

1999.
340
(8):603–8.

the etiology of schizophrenia.

21. Buka S. L., Tsuang M. T., Torrey

4. Bagalkote H. Maternal influenza

Arch Gen Psychiatry, 2004.

E. F.,
et al.
Maternal infections

and schizophrenia in the

61
:774–80.

and subsequent psychosis among

offspring. Int J Ment Health, 2001.

14. DeLisi L. E. Is there a viral or

offspring. Arch Gen Psychiatry,

29
(4):3–21.

immune dysfunction etiology to

2001.
58
(11):1032–7.

5. Brown A. S., Susser E. S. In utero

schizophrenia? Re-evaluation a

22. Brown A. S., Schaefer C. A.,

infection and adult schizophrenia.

decade later. Schizophr Res, 1996.

Quesenberry C. P. Jr., Shen L.,

Ment Retard Dev Disabil Res Rev,

22
(1):1–4.

Susser E. S. No evidence of

2002.
8
(1):51–7.

15. Wright P., Takei N., Murray R. M.,

relation between maternal

6. Watson C. G., Kucala T., Tilleskjor

Sham P. C. (1999). Seasonality,

exposure to herpes simplex virus

C., Jacobs L. Schizophrenic birth

prenatal influenza exposure, and

type 2 and risk of schizophrenia.

seasonality in relation to the

schizophrenia. In Prenatal

Am J Psychiatry, 2006.
163
(12):
incidence of infectious diseases

Exposures in Schizophrenia, Susser

2178–80.

and temperature extremes. Arch

E. S., Brown A. S., and Gorman

23. Corey L., Whitley R. J., Stone

Gen Psychiatry, 1984.
41
(1):
J. M. (Eds.). Washington, DC: