. 34
( 68 .)



45 44


Manic Mixed affective Schizodepressive Schizophrenic
(n = 198) (n = 46) (n = 92) (n = 101)

Depressive Schizomanic Mixed schizoaffective
(n = 249) (n = 168) (n = 53)
Fig. 8.4 Duration of episodes (length of hospitalization).

literature on the topic, the longest-lasting episodes (Kraepelin, 1899; Weygandt,
1899; Marneros et al., 1991).
The findings of HABILOS showed no significant differences in episode length
between different episode types (Fig. 8.4). On a purely descriptive level, the mixed
schizoaffective episodes had the longest duration of all types of episode. Descriptively,
mixed affective episodes had almost the same duration as pure manic episodes “
the shortest of all. These findings are contrary to our previous findings (Marneros
et al., 1991). The explanation we have is that, while almost none of the patients with
mixed episodes in the Cologne study received anticonvulsants as mood stabilizers,
in contrast the vast majority of the HABILOS patients with mixed episodes received
198 A. Marneros et al.

Bipolar affective Bipolar schizoaffective
(n = 100) (n = 177)
Mixed Mixed
25.0% 32.2%

Non-mixed Non-mixed
75.0% 67.8%
Fig. 8.5 Frequency of patients with at least one mixed episode.

anticonvulsants, especially valproate, carbamazepine, and lamotrigine. The positive
response of mixed states to the above-mentioned mood stabilizer have been repeat-
edly documented (Calabrese et al., 2000; see Chapter 3). Perhaps this is an effect
of the new mood stabilizers. Additionally, none of the patients in the Cologne
study received atypical neuroleptics because the study was closed before atypical
neuroleptics were introduced, while almost all patients with mixed states of the
HABILOS received atypical neuroleptics. The positive response of patients with
mixed states to typical neuroleptics is also documented (Tohen, 2000; see
chapter 16). Perhaps treatment with atypical neuroleptics or even a combination
of novel neuroleptics with novel mood stabilizers, is another factor contributing
to the shortness of episodes. Finally, one should not overemphasize these find-
ings, as statistically the length of duration of treatment was identical for all
episode types. More indepth analysis will be needed.

Frequency of patients having mixed episodes
Following the ˜˜empirical longitudinal diagnosis of schizoaffective disorder™™
(Marneros et al., 1986, 1991), we were able to allocate all the HABILOS patients
into two categories:
(1) bipolar affective patients, who had only affective episodes (depressive, manic,
mixed) but no schizoaffective or schizophrenic episodes during the entire
course of their illness
(2) bipolar schizoaffective patients, with at least one schizoaffective episode dur-
ing the illness course, or when schizophrenic and affective episodes occured
sequentially (see Chapter 1)
Applying this clinical and longitudinal classification, 100 patients were diagnosed
as bipolar affective and 177 as bipolar schizoaffective. The high proportion of
schizoaffective courses may reflect the specialization of our hospital in such dis-
orders. As shown in Figure 8.5, 25% of the bipolar affective patients had at least
199 Schizoaffective mixed states

Female Male

Total bipolar (n = 277) 48.7% 51.3%

Bipolar overall not mixed 46.7% 53.3%
(n = 195)

Mixed overall (n = 82) 53.7% 46.3%

Bipolar affective not 52.0% 48.0%
mixed (n = 75)

Bipolar affective mixed 60.0% 40.0%
(n = 25)

43.3% 46.7%
Bipolar schizoaffective
not mixed (n = 120)

Bipolar schizoaffective 49.1%
mixed (n = 57)
60% 40% 20% 0% 20% 40% 60%

Fig. 8.6 Gender distribution.

one mixed episode, while 32.2% of the bipolar schizoaffective patients had at least
one mixed episode (not statistically significant difference).

Gender distribution
The findings of the HABILOS confirmed the findings of former investigations,
that females are overrepresented in the group of affective mixed states (McElroy
et al., 2000; Marneros et al., 1991), although this difference did not reach signi-
ficance. The overrepresentation of females in the group of mixed states, however,
was only found in the group of pure affective mixed states, but not in that of
schizoaffective mixed, where gender is almost equally represented. This is possibly
a result of the impact of the schizophrenic symptomatology. It is well known that
in schizophrenia the percentage of males is equal to that of females, if not higher.
Taking into consideration that the gender distribution in the whole bipolar group
of the HABILOS is almost equal “ as was expected “ we can conclude that the
findings regarding the gender distribution in the subgroups (like mixed states)
could also be assumed to be representative (Fig. 8.6).

Initial episode and first manifestation of a mixed episode
Only 14% of ˜˜schizoaffective patients™™ and 20% of ˜˜mixed pure affective patients™™
had a mixed initial episode (Table 8.5).
Mixed episodes can manifest any time during the course of the illness.
Nevertheless, more than 40% of all patients with mixed episodes had their first
mixed episode after the fourth illness episode (Fig. 8.7).
200 A. Marneros et al.

Table 8.5 Types of initial episode

Bipolar affective disorders Bipolar schizoaffective disorders
(n ¼ 100) (n ¼ 177)

Non-mixed Mixed Non-mixed Mixed
(n ¼ 75) (n ¼ 25) (n ¼ 120) (n ¼ 57)

Manic 28 (37.3%) 6 (24.0%) 13 (10.8%) 2 (3.5%)
Depressive 45 (60.0%) 13 (52.0%) 25 (20.8%) 21 (36.8%)
Mixed affective 5 (20.0%)
Schizomanic 29 (24.2%) 6 (10.5%)
Schizodepressive 7 (5.8%) 1 (1.8%)
Mixed schizoaffective 8 (14.0%)
Schizophrenic 41 (34.2%) 13 (22.8%)
Not exactly 2 (27%) 1 (4.0%) 5 (4.2%) 6 (10.5%)

1st episode
2nd episode

3rd episode

4th episode
5th episode

After 5th episode

0% 10% 20% 30% 40% 50%

Bipolar affective Bipolar schizoaffective

Fig. 8.7 Time of first manifestation of a mixed episode.

Disability pensions
In both groups “ bipolar affective and bipolar schizoaffective disorders “ patients
having mixed episodes are overrepresented in receiving disability payments due to
their illness. This is especially the case in the mixed schizoaffective group, with almost
60% of the patients receiving disability pensions. But also in the group of pure
affective mixed episodes, a rate of 52% seems rather high. This means that patients
with mixed schizoaffective episodes have a less favorable prognosis and more
201 Schizoaffective mixed states

Bipolar disorders (n = 277)




Bipolar affective Bipolar schizoaffective

Non-mixed (n = 75) Mixed (n = 25) Non-mixed (n = 120) Mixed (n = 57)
no no no no
48.0% 47.5%

52.0% 59.6%

yes yes yes yes

Fig. 8.8 Patients receiving disability pensions.

Analysis of difference
1. Bipolar affective non-mixed versus
60 NS
bipolar affective mixed
2. Bipolar SAD non-mixed versus
50 bipolar SAD mixed NS
3. Bipolar mixed versus bipolar SAD
40 36.0
4. Bipolar affective non-mixed versus
bipolar schizoaffective non-
P = 0.001


Bipolar affective Bipolar SAD non-mixed
non-mixed (n = 63)
(n = 26)
Bipolar affective Bipolar SAD mixed
mixed (n = 34)
(n = 13)

Fig. 8.9 Age when disability payments were first granted. NS, not significant; SAD, seasonal affective

disability. As Figure 8.8 shows, at the time of investigation, half of the patients with
bipolar disorder were receiving disability pensions as a result of their mental illness.
In the group of schizoaffective mixed and non-mixed patients, disability pen-
sions were granted at the age of approximately 36 years “ a very low age.
202 A. Marneros et al.

In the state of Saxony-Anhalt (Germany), where this study was carried out, all


. 34
( 68 .)