|
Numerous sex differences have been reported in schizophrenia, and
it is now evident that men and women experience and manifest psychosis
differently. Briefly, women show a superior premorbid adjustment,
relative to men: their symptoms begin later in life, and outcome
is usually superior (at least in the first 15 years after onset)
(1). Women also show a second paramenopausal peak onset not seen
in men (2). Mood symptoms such as depression are more common in
women, whereas apathy, flat affect, paucity of speech, and social
isolation are more often seen in men (3), although this difference
in symptoms is not always seen (1). More brain structure impairment
has been reported in men (4). Premenopausal women may respond at
lower antipsychotic dosages than do men (5). Although the etiology
of the sex differences is not entirely understood, hormone-gene
interactions probably account for part of the variance. The first
part of this paper reviews the literature on studies examining the
influence of estrogen in schizophrenia. This review is based on
a Medline database search and a review of relevant reference lists.
To optimize diagnosis and treatment of schizophrenia in women, the
second half of this paper suggests modifications of practice guidelines,
based on the evidence of estrogen effects. To date, guidelines for
the treatment of schizophrenia have not directly distinguished between
the sexes.
The association between estrogen and schizophrenia in women is
not a 20th century phenomenon. Early clinicians referred to many
women with schizophrenia as suffering from “hypoestrogenism.”
Schizophrenic psychosis was said to be influenced by the natural
variation of estrogen levels, both over a woman’s cycle and
over her lifetime (6). The early observations are especially important
because they were made before the advent of modern antipsychotics
that, via dopamine blockade and subsequent release of prolactin
inhibition, result in high prolactin serum levels with downstream
inhibition of ovarian estrogen secretion; that is, secondary hypo-estrogenism.
The modern “estrogen hypothesis” postulates that estrogen
exerts a protective effect against schizophrenia and that this partly
explains the observed sex differences in premorbid adjustment, onset
age, treatment response, and illness course. Evidence for the role
of estrogen in schizophrenia comes from many sources, including
epidemiologic data, effects of pubertal age on schizophrenia onset,
the changing severity of symptoms over a woman’s reproductive
life, and the results of treatment studies.
Animal studies show estrogens to have organizational effects on
developing neurons and activational effects on mature neurons. Estrogens
affect neurite growth and synapse formation; interact with nerve
growth factor and other neurotrophins; and modulate many neurotransmitters
systems, including the dopamine, serotonin, norepinephrine, acetylcholine,
and glutamate systems. Table
1 presents estrogens’ multiple other protective effects
(7).
Several studies have established that schizophrenia has later onset
in women. Typically, women present at age 25 to 29 years, as opposed
to men, who usually first present at age 20 to 24 years (2). In
addition, a second smaller peak of onset exists for women after
age 44 years, around the perimenopause and menopause (8).
Earlier age at menarche is associated with later onset of schizophrenic
symptoms (9). In contrast to what happens after puberty, prepubertal
onset is earlier in girls than in boys (10). These 2 findings taken
together suggest that pubertal hormones delay onset in women. At
the other end of life, symptom severity increases with age in women
only (11).
Forty-seven percent of psychiatric admissions occur during the
paramenstrual phase, when estrogen levels are lowest (12). Hallonquist
and others looked at 5 outpatients over 2 consecutive menstrual
cycles and found global symptom scores to be significantly lower
during high estrogen phases (13).
|
|
A more recent study with a larger sample confirmed these findings
(14). Riecher-Rossler and others showed a significant association
between estradiol levels and psychosis scores, with psychopathology
improving when estradiol levels rose (15). Hoff and others studied
a group of 22 inpatients and found a positive association between
estrogen levels and cognitive performance (16). Taken together,
these studies suggest a modulating effect of estrogen on both symptoms
and cognition: the higher the estrogen levels, the less severe the
symptoms, and the more intact the cognitive abilities.
Chang and Renshaw reported an amelioration of symptoms during pregnancy
(17). Krener and others found that women with schizophrenia received
and required less medication to control their thought disorder when
they were pregnant (18). Kendall and others noted that relapses
tended to occur postpartum, when estrogen levels were low (19).
One study showed that women with schizophrenia admitted during
the low-estrogen phases of their menstrual cycle required significantly
lower mean daily dosages of neuroleptic medication (325 mg chlorpromazine
[CPZ], SD 203 mg) than did a high-estrogen phase admission group
(414 mg CPZ , SD 204 mg; P < 0.05) (20). From this study,
Gattaz and others concluded that the antidopaminergic effects of
estradiol modulate the vulnerability threshold for psychosis. In
a 3-year survey of the neuroleptic maintenance requirements of 101
outpatients suffering from chronic schizophrenia, Seeman found that
younger women (age 20 to 39 years) were maintained on lower dosages
than were men, but the reverse was true after age 40 years (5).
Not all studies agree, however. Although Szymanski and others found
a greater pharmacologic response in female patients with first-episode
schizophrenia, compared with male patients in one study (21), treatment-refractory
women did not fare better then men on clozapine in a subsequent
study (22).
Direct evidence for the estrogen hypothesis comes from studies
that have used estrogen to treat psychotic symptoms. In a 28-day
double-blind randomized controlled trial (RCT), Kulkarni and others
found dramatic improvement in psychotic symptoms beginning at day
4 (23). In this study, 36 women were randomly assigned either a
100 mcg estrogen patch, a 50 mcg estrogen patch, or placebo, in
addition to standardized antipsychotic medication. The women in
the 100 mcg group were significantly better at the end of the trial.
In an earlier open clinical trial, Kulkarni and others gave 0.02
mg of estradiol in addition to regular neuroleptic treatment to
11 women with acute psychotic symptoms (24). Their response was
compared with that of 7 women with similar symptom severity receiving
neuroleptic treatment alone. The group receiving the estradiol adjunct
showed more rapid improvement in psychotic symptoms, although both
groups reached similar levels of recovery by the final week of the
study (week 8). Lindamer and others presented a case report of estradiol
augmentation in a postmenopausal woman with schizophrenia (25).
They administered 0.05 mg of Estraderm in addition to perphenazine
and found a reduction in the patient’s positive symptoms while
on estrogen. Once the patient discontinued the estrogen, she returned
to near-baseline levels of positive symptoms. Korhonen reported
that estradiol therapy alone alleviated psychotic symptoms occurring
in the premenstrual phase of a woman diagnosed with schizoaffective
disorder (26). This patient was given 3.0 mg of percutaneous estradiol
for 5 months. She was able to discontinue her regular medications
1 month after starting estradiol. Taken together, the results of
these studies suggest that estradiol demonstrates antipsychotic
properties or acts as a catalyst for neuroleptic responsiveness
(or both) in women with schizophrenia. Similarly, Ahokas and others
were able to successfully treat 10 women with postpartum psychosis
with 17ß-estradiol alone (27). Scores on the Brief Psychiatric
Rating Scale improved from a mean of 78 (SD 7.7) to 19 (SD 6.9)
after the first week of treatment. This study also showed the link
between low serum estradiol concentrations and clinical response
to estradiol treatment. Although the women in the study suffered
from postpartum psychosis and not schizophrenia, estradiol was shown
to have a direct antipsychotic effect.
|
